vjev-vision: full vision run, step 600 - merged bf16 model, head, render config, stand-alone inference
Browse files- .gitattributes +1 -0
- README.md +114 -0
- adapter/adapter_config.json +63 -0
- adapter/adapter_model.safetensors +3 -0
- chat_template.jinja +154 -0
- config.json +113 -0
- generation_config.json +6 -0
- head.pt +3 -0
- model-00001-of-00003.safetensors +3 -0
- model-00002-of-00003.safetensors +3 -0
- model-00003-of-00003.safetensors +3 -0
- model.safetensors.index.json +731 -0
- processor_config.json +61 -0
- tokenizer.json +3 -0
- tokenizer_config.json +33 -0
- vjev.json +10 -0
- vjev_infer.py +178 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,114 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen3.5-4B
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
library_name: transformers
|
| 5 |
+
pipeline_tag: image-text-to-text
|
| 6 |
+
tags:
|
| 7 |
+
- qwen3.5
|
| 8 |
+
- classification
|
| 9 |
+
- calibration
|
| 10 |
+
- vision
|
| 11 |
+
- jev
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# vjev-vision
|
| 15 |
+
|
| 16 |
+
A **listwise decision model** with vision: give it a state (text, images, or both) and typed
|
| 17 |
+
questions — `noul` (is this statement true?), `choice` (pick one), `score` (an ordered scale) —
|
| 18 |
+
and it returns calibrated probabilities for every option, in a single forward pass, with no
|
| 19 |
+
text generation. It re-creates the Jev API's shape on an open base (Qwen3.5-4B), with images added.
|
| 20 |
+
|
| 21 |
+
This is the released checkpoint of the full vision run (step 600 of 1,200; chosen over the final
|
| 22 |
+
step for its yes/no behaviour, see below). The earlier 300-step pilot stays at
|
| 23 |
+
[yah01/vjev-vision-pilot](https://huggingface.co/yah01/vjev-vision-pilot).
|
| 24 |
+
|
| 25 |
+
## Use it
|
| 26 |
+
|
| 27 |
+
The web console and the HTTP API live in [BubbleCal/vjev-serve](https://github.com/BubbleCal/vjev-serve):
|
| 28 |
+
|
| 29 |
+
```bash
|
| 30 |
+
pip install git+https://github.com/BubbleCal/vjev-serve
|
| 31 |
+
vjev-serve --model yah01/vjev-vision # then open http://localhost:8800
|
| 32 |
+
```
|
| 33 |
+
|
| 34 |
+
Or, with nothing but transformers, the single file in this repo:
|
| 35 |
+
|
| 36 |
+
```python
|
| 37 |
+
from huggingface_hub import hf_hub_download
|
| 38 |
+
import importlib.util
|
| 39 |
+
spec = importlib.util.spec_from_file_location("vjev_infer", hf_hub_download("yah01/vjev-vision", "vjev_infer.py"))
|
| 40 |
+
vjev_infer = importlib.util.module_from_spec(spec); spec.loader.exec_module(vjev_infer)
|
| 41 |
+
|
| 42 |
+
m = vjev_infer.Vjev("yah01/vjev-vision") # cuda / mps / cpu
|
| 43 |
+
m.ask(state=["photo.jpg", "Frame from the warehouse camera, 12:40."],
|
| 44 |
+
questions={"person": {"type": "noul", "instructions": "There is a person in this image."},
|
| 45 |
+
"where": {"type": "choice", "instructions": "Where is the forklift?",
|
| 46 |
+
"criteria": {"left": "left half", "right": "right half", "none": "no forklift"}},
|
| 47 |
+
"busy": {"type": "score", "instructions": "How cluttered is the scene?",
|
| 48 |
+
"criteria": ["empty", "sparse", "busy", "crowded"]}})
|
| 49 |
+
```
|
| 50 |
+
|
| 51 |
+
Options inside one question compete (one softmax); questions never see each other. To rate
|
| 52 |
+
several independent things, ask one `noul` per thing (vjev-serve's `multilabel` type does this).
|
| 53 |
+
About 9 GB of memory in bf16/fp16.
|
| 54 |
+
|
| 55 |
+
## What is in this repo
|
| 56 |
+
|
| 57 |
+
| file | what |
|
| 58 |
+
|---|---|
|
| 59 |
+
| `model-*.safetensors`, `config.json`, tokenizer and processor files | the full model, bf16: Qwen3.5-4B with the checkpoint's LoRA merged in |
|
| 60 |
+
| `head.pt` | the listwise scoring head: one shared linear layer read at each option's slot |
|
| 61 |
+
| `vjev.json` | how inputs are rendered (`readout: trailing`, `pause: 0`, length budgets) |
|
| 62 |
+
| `vjev_infer.py` | stand-alone inference |
|
| 63 |
+
| `adapter/` | the same weights as a LoRA adapter (PEFT, r=32, α=64) over `Qwen/Qwen3.5-4B`, fitted on the nf4-quantized base |
|
| 64 |
+
|
| 65 |
+
The merged weights are the nf4-dequantized base the adapter was trained against, plus the
|
| 66 |
+
adapter, in bf16 — no bitsandbytes needed. Against the adapter on its nf4 base (CUDA, 118
|
| 67 |
+
questions over 10 images): max |Δp| 0.0097, mean 0.0036, no answer changed.
|
| 68 |
+
|
| 69 |
+
## How it was trained
|
| 70 |
+
|
| 71 |
+
1. **Text** (3,868 steps): QLoRA on ~145k typed questions with soft labels from the Jev API plus
|
| 72 |
+
human-labelled corpora; KL to the teacher distribution for choice/score, soft BCE for noul.
|
| 73 |
+
All options of a question share one sequence and are read at a trailing `Answer: (A) (B) …`
|
| 74 |
+
slot, so options can see each other.
|
| 75 |
+
2. **Vision** (this checkpoint, step 600 of a 1,200-step run warm-started from the text stage):
|
| 76 |
+
a second LoRA on the vision tower (0.1× the text learning rate), on ~84k geometry questions
|
| 77 |
+
from COCO-2017 annotations (which object is highest / smallest / left of…, counts, presence
|
| 78 |
+
with adversarial absent objects; soft labels by Monte-Carlo perturbation of the boxes) and
|
| 79 |
+
~15k VQAv2 questions with their 10-annotator answer distributions, mixed with 25% text.
|
| 80 |
+
|
| 81 |
+
## Results (held out: images and texts never seen in training)
|
| 82 |
+
|
| 83 |
+
| | text stage (zero-shot) | pilot, 300 steps | **this, step 600** | final step 1200 |
|
| 84 |
+
|---|---:|---:|---:|---:|
|
| 85 |
+
| COCO geometry, choice accuracy | 0.506 | 0.665 | **0.735** | 0.703 |
|
| 86 |
+
| COCO geometry, choice ECE ↓ | 0.171 | 0.094 | 0.183 | 0.148 |
|
| 87 |
+
| VQAv2, choice accuracy | 0.579 | 0.654 | 0.706 | 0.721 |
|
| 88 |
+
| VQAv2, choice ECE ↓ | 0.126 | 0.033 | 0.052 | 0.070 |
|
| 89 |
+
| POPE adversarial, absent objects called present ↓ | 5.5% | 13.8% | **6.1%** | 12.9% |
|
| 90 |
+
| POPE, AUC | 0.977 | 0.969 | 0.966 | 0.962 |
|
| 91 |
+
| text, 20k held-out rows: accuracy vs human labels | 0.796 | — | 0.799 | 0.797 |
|
| 92 |
+
| text, 20k held-out rows: ECE ↓ | 0.068 | — | 0.063 | 0.065 |
|
| 93 |
+
|
| 94 |
+
Step 600 was picked over the final step: the final step called absent objects present twice
|
| 95 |
+
as often, for a small VQA gain. On text the vision stage cost nothing; the teacher itself scores
|
| 96 |
+
0.797 accuracy and 0.102 ECE on the same rows.
|
| 97 |
+
|
| 98 |
+
Options interact as in the teacher: adding a competing option protects the leader and takes
|
| 99 |
+
its mass from the runners-up (a paired effect of +0.0066 over the order-permutation noise
|
| 100 |
+
floor, CI excluding 0) — something a one-option-per-pass scorer cannot do.
|
| 101 |
+
|
| 102 |
+
## Limits
|
| 103 |
+
|
| 104 |
+
- Spatial answers are over-confident (ECE 0.18 on COCO geometry): the ranking is better than
|
| 105 |
+
the probabilities.
|
| 106 |
+
- Trained on single images; several images in one request (`Picture 1:`, `Picture 2:`) rely on
|
| 107 |
+
the base model's ability.
|
| 108 |
+
- Options are read in the order given; reordering them moves probabilities by a TVD of ~0.037
|
| 109 |
+
on average.
|
| 110 |
+
- Images are resized to a 512 px longer side, the training resolution.
|
| 111 |
+
|
| 112 |
+
## License
|
| 113 |
+
|
| 114 |
+
Apache-2.0. The base model, Qwen3.5-4B, is Apache-2.0.
|
adapter/adapter_config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": {
|
| 6 |
+
"base_model_class": "Qwen3_5Model",
|
| 7 |
+
"parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
|
| 8 |
+
},
|
| 9 |
+
"base_model_name_or_path": "Qwen/Qwen3.5-4B",
|
| 10 |
+
"bias": "none",
|
| 11 |
+
"corda_config": null,
|
| 12 |
+
"ensure_weight_tying": false,
|
| 13 |
+
"eva_config": null,
|
| 14 |
+
"exclude_modules": null,
|
| 15 |
+
"fan_in_fan_out": false,
|
| 16 |
+
"inference_mode": true,
|
| 17 |
+
"init_lora_weights": true,
|
| 18 |
+
"kasa_config": null,
|
| 19 |
+
"layer_replication": null,
|
| 20 |
+
"layers_pattern": null,
|
| 21 |
+
"layers_to_transform": null,
|
| 22 |
+
"loftq_config": {},
|
| 23 |
+
"lora_alpha": 64,
|
| 24 |
+
"lora_bias": false,
|
| 25 |
+
"lora_dropout": 0.05,
|
| 26 |
+
"lora_ga_config": null,
|
| 27 |
+
"megatron_config": null,
|
| 28 |
+
"megatron_core": "megatron.core",
|
| 29 |
+
"modules_to_save": null,
|
| 30 |
+
"monteclora_config": null,
|
| 31 |
+
"peft_type": "LORA",
|
| 32 |
+
"peft_version": "0.21.0",
|
| 33 |
+
"qalora_group_size": 16,
|
| 34 |
+
"r": 32,
|
| 35 |
+
"rank_pattern": {},
|
| 36 |
+
"revision": null,
|
| 37 |
+
"target_modules": [
|
| 38 |
+
"in_proj_b",
|
| 39 |
+
"in_proj_z",
|
| 40 |
+
"linear_fc1",
|
| 41 |
+
"down_proj",
|
| 42 |
+
"in_proj_qkv",
|
| 43 |
+
"k_proj",
|
| 44 |
+
"in_proj_a",
|
| 45 |
+
"o_proj",
|
| 46 |
+
"linear_fc2",
|
| 47 |
+
"q_proj",
|
| 48 |
+
"up_proj",
|
| 49 |
+
"v_proj",
|
| 50 |
+
"out_proj",
|
| 51 |
+
"qkv",
|
| 52 |
+
"attn.proj",
|
| 53 |
+
"gate_proj"
|
| 54 |
+
],
|
| 55 |
+
"target_parameters": null,
|
| 56 |
+
"task_type": null,
|
| 57 |
+
"trainable_token_indices": null,
|
| 58 |
+
"use_bdlora": null,
|
| 59 |
+
"use_dora": false,
|
| 60 |
+
"use_qalora": false,
|
| 61 |
+
"use_rslora": false,
|
| 62 |
+
"velora_config": null
|
| 63 |
+
}
|
adapter/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f31d09d8b721d5c105708f208d5cea3208077743084727291b793327f8145b06
|
| 3 |
+
size 312049688
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "bfloat16",
|
| 6 |
+
"image_token_id": 248056,
|
| 7 |
+
"model_type": "qwen3_5",
|
| 8 |
+
"text_config": {
|
| 9 |
+
"attention_bias": false,
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"attn_output_gate": true,
|
| 12 |
+
"bos_token_id": null,
|
| 13 |
+
"dtype": "bfloat16",
|
| 14 |
+
"eos_token_id": 248044,
|
| 15 |
+
"full_attention_interval": 4,
|
| 16 |
+
"head_dim": 256,
|
| 17 |
+
"hidden_act": "silu",
|
| 18 |
+
"hidden_size": 2560,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 9216,
|
| 21 |
+
"layer_types": [
|
| 22 |
+
"linear_attention",
|
| 23 |
+
"linear_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"linear_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"linear_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"linear_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"linear_attention",
|
| 53 |
+
"full_attention"
|
| 54 |
+
],
|
| 55 |
+
"linear_conv_kernel_dim": 4,
|
| 56 |
+
"linear_key_head_dim": 128,
|
| 57 |
+
"linear_num_key_heads": 16,
|
| 58 |
+
"linear_num_value_heads": 32,
|
| 59 |
+
"linear_value_head_dim": 128,
|
| 60 |
+
"mamba_ssm_dtype": "float32",
|
| 61 |
+
"max_position_embeddings": 262144,
|
| 62 |
+
"mlp_only_layers": [],
|
| 63 |
+
"model_type": "qwen3_5_text",
|
| 64 |
+
"mtp_num_hidden_layers": 1,
|
| 65 |
+
"mtp_use_dedicated_embeddings": false,
|
| 66 |
+
"num_attention_heads": 16,
|
| 67 |
+
"num_hidden_layers": 32,
|
| 68 |
+
"num_key_value_heads": 4,
|
| 69 |
+
"pad_token_id": null,
|
| 70 |
+
"partial_rotary_factor": 0.25,
|
| 71 |
+
"rms_norm_eps": 1e-06,
|
| 72 |
+
"rope_parameters": {
|
| 73 |
+
"mrope_interleaved": true,
|
| 74 |
+
"mrope_section": [
|
| 75 |
+
11,
|
| 76 |
+
11,
|
| 77 |
+
10
|
| 78 |
+
],
|
| 79 |
+
"partial_rotary_factor": 0.25,
|
| 80 |
+
"rope_theta": 10000000,
|
| 81 |
+
"rope_type": "default"
|
| 82 |
+
},
|
| 83 |
+
"tie_word_embeddings": true,
|
| 84 |
+
"use_cache": true,
|
| 85 |
+
"vocab_size": 248320
|
| 86 |
+
},
|
| 87 |
+
"tie_word_embeddings": true,
|
| 88 |
+
"transformers_version": "5.17.0",
|
| 89 |
+
"video_token_id": 248057,
|
| 90 |
+
"vision_config": {
|
| 91 |
+
"deepstack_visual_indexes": [],
|
| 92 |
+
"depth": 24,
|
| 93 |
+
"dtype": "bfloat16",
|
| 94 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 95 |
+
"hidden_size": 1024,
|
| 96 |
+
"in_channels": 3,
|
| 97 |
+
"initializer_range": 0.02,
|
| 98 |
+
"intermediate_size": 4096,
|
| 99 |
+
"model_type": "qwen3_5_vision",
|
| 100 |
+
"num_heads": 16,
|
| 101 |
+
"num_position_embeddings": 2304,
|
| 102 |
+
"out_hidden_size": 2560,
|
| 103 |
+
"patch_size": 16,
|
| 104 |
+
"rope_parameters": {
|
| 105 |
+
"rope_theta": 10000.0,
|
| 106 |
+
"rope_type": "axial"
|
| 107 |
+
},
|
| 108 |
+
"spatial_merge_size": 2,
|
| 109 |
+
"temporal_patch_size": 2
|
| 110 |
+
},
|
| 111 |
+
"vision_end_token_id": 248054,
|
| 112 |
+
"vision_start_token_id": 248053
|
| 113 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": 248044,
|
| 4 |
+
"transformers_version": "5.17.0",
|
| 5 |
+
"use_cache": true
|
| 6 |
+
}
|
head.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1c21ff4eee514b109ab6299c8c76e5e8cf7925ba14141fe18d3d930a92a2629
|
| 3 |
+
size 12045
|
model-00001-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24ba02d1fb2ce94b82dbb5392bff0ad0a6da68af29de91b20069094523686daf
|
| 3 |
+
size 3991298872
|
model-00002-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1a174029ad8264cbd2aa8cb701321cd432ddf020a2da910d526365c49b8f071
|
| 3 |
+
size 3979833152
|
model-00003-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:611359f82c25f0c13abfdcf115812ca23ffd6765d7bbd330ff2d6631756b4f71
|
| 3 |
+
size 1107487880
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,731 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 4539265536,
|
| 4 |
+
"total_size": 9078531072
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00003.safetensors",
|
| 8 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 9 |
+
"model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 17 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 18 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 20 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 21 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 22 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 23 |
+
"model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 31 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 32 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 34 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 35 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 36 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 37 |
+
"model.language_model.layers.10.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 45 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 46 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 48 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 49 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 50 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 51 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 53 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 54 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 55 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 60 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 61 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 62 |
+
"model.language_model.layers.12.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 70 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 71 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 73 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 74 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 75 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 76 |
+
"model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 84 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 85 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 87 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 88 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 89 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 90 |
+
"model.language_model.layers.14.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 98 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 99 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 101 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 102 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 103 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 104 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 106 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 107 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 108 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 113 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 114 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 115 |
+
"model.language_model.layers.16.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 123 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 124 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 126 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 127 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 128 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 129 |
+
"model.language_model.layers.17.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 137 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 138 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 140 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 141 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 142 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 143 |
+
"model.language_model.layers.18.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 151 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 152 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 154 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 155 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 156 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 157 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 159 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 160 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 161 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 166 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 167 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 168 |
+
"model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 176 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 177 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 179 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 180 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 181 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 182 |
+
"model.language_model.layers.20.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 190 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 191 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 193 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 194 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 195 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 196 |
+
"model.language_model.layers.21.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 204 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 205 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 207 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 208 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 209 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 210 |
+
"model.language_model.layers.22.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 218 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 219 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 221 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 222 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 223 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 224 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 226 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 227 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 228 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 233 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 234 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 235 |
+
"model.language_model.layers.24.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 243 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 244 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 246 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 247 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 248 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 249 |
+
"model.language_model.layers.25.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 257 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 258 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 260 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 261 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 262 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 263 |
+
"model.language_model.layers.26.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 271 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 272 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 274 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 275 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 276 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 277 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 279 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 280 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 281 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00002-of-00003.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00002-of-00003.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 286 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 287 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 288 |
+
"model.language_model.layers.28.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 296 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 297 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 299 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 300 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 301 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 302 |
+
"model.language_model.layers.29.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00002-of-00003.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00002-of-00003.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model-00002-of-00003.safetensors",
|
| 310 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model-00002-of-00003.safetensors",
|
| 311 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 313 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 314 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 315 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 316 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 318 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 319 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 320 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 325 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 326 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 327 |
+
"model.language_model.layers.30.linear_attn.A_log": "model-00002-of-00003.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model-00002-of-00003.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model-00002-of-00003.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00002-of-00003.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00002-of-00003.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00003-of-00003.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00003-of-00003.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model-00003-of-00003.safetensors",
|
| 335 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model-00003-of-00003.safetensors",
|
| 336 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 338 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 339 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 340 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 341 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 343 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 344 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 345 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00003-of-00003.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00003-of-00003.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 350 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 351 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 352 |
+
"model.language_model.layers.4.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 360 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 361 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 363 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 364 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 365 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 366 |
+
"model.language_model.layers.5.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 374 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 375 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 377 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 378 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 379 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 380 |
+
"model.language_model.layers.6.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 388 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 389 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 391 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 392 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 393 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 394 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 396 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 397 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 398 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00003.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00003.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 403 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 404 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 405 |
+
"model.language_model.layers.8.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 413 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 414 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 416 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 417 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 418 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 419 |
+
"model.language_model.layers.9.linear_attn.A_log": "model-00001-of-00003.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model-00001-of-00003.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model-00001-of-00003.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00001-of-00003.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00001-of-00003.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00001-of-00003.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00001-of-00003.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model-00001-of-00003.safetensors",
|
| 427 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 428 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 430 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 431 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 432 |
+
"model.language_model.norm.weight": "model-00003-of-00003.safetensors",
|
| 433 |
+
"model.visual.blocks.0.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 436 |
+
"model.visual.blocks.0.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 437 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 440 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 441 |
+
"model.visual.blocks.0.norm1.bias": "model-00003-of-00003.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm1.weight": "model-00003-of-00003.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm2.bias": "model-00003-of-00003.safetensors",
|
| 444 |
+
"model.visual.blocks.0.norm2.weight": "model-00003-of-00003.safetensors",
|
| 445 |
+
"model.visual.blocks.1.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 448 |
+
"model.visual.blocks.1.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 449 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 452 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 453 |
+
"model.visual.blocks.1.norm1.bias": "model-00003-of-00003.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm1.weight": "model-00003-of-00003.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm2.bias": "model-00003-of-00003.safetensors",
|
| 456 |
+
"model.visual.blocks.1.norm2.weight": "model-00003-of-00003.safetensors",
|
| 457 |
+
"model.visual.blocks.10.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 460 |
+
"model.visual.blocks.10.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 461 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 464 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 465 |
+
"model.visual.blocks.10.norm1.bias": "model-00003-of-00003.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm1.weight": "model-00003-of-00003.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm2.bias": "model-00003-of-00003.safetensors",
|
| 468 |
+
"model.visual.blocks.10.norm2.weight": "model-00003-of-00003.safetensors",
|
| 469 |
+
"model.visual.blocks.11.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 472 |
+
"model.visual.blocks.11.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 473 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 476 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 477 |
+
"model.visual.blocks.11.norm1.bias": "model-00003-of-00003.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm1.weight": "model-00003-of-00003.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm2.bias": "model-00003-of-00003.safetensors",
|
| 480 |
+
"model.visual.blocks.11.norm2.weight": "model-00003-of-00003.safetensors",
|
| 481 |
+
"model.visual.blocks.12.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 484 |
+
"model.visual.blocks.12.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 485 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 488 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 489 |
+
"model.visual.blocks.12.norm1.bias": "model-00003-of-00003.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm1.weight": "model-00003-of-00003.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm2.bias": "model-00003-of-00003.safetensors",
|
| 492 |
+
"model.visual.blocks.12.norm2.weight": "model-00003-of-00003.safetensors",
|
| 493 |
+
"model.visual.blocks.13.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 496 |
+
"model.visual.blocks.13.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 497 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 500 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 501 |
+
"model.visual.blocks.13.norm1.bias": "model-00003-of-00003.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm1.weight": "model-00003-of-00003.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm2.bias": "model-00003-of-00003.safetensors",
|
| 504 |
+
"model.visual.blocks.13.norm2.weight": "model-00003-of-00003.safetensors",
|
| 505 |
+
"model.visual.blocks.14.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 508 |
+
"model.visual.blocks.14.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 509 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 512 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 513 |
+
"model.visual.blocks.14.norm1.bias": "model-00003-of-00003.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm1.weight": "model-00003-of-00003.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm2.bias": "model-00003-of-00003.safetensors",
|
| 516 |
+
"model.visual.blocks.14.norm2.weight": "model-00003-of-00003.safetensors",
|
| 517 |
+
"model.visual.blocks.15.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 520 |
+
"model.visual.blocks.15.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 521 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 524 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 525 |
+
"model.visual.blocks.15.norm1.bias": "model-00003-of-00003.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm1.weight": "model-00003-of-00003.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm2.bias": "model-00003-of-00003.safetensors",
|
| 528 |
+
"model.visual.blocks.15.norm2.weight": "model-00003-of-00003.safetensors",
|
| 529 |
+
"model.visual.blocks.16.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 532 |
+
"model.visual.blocks.16.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 533 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 536 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 537 |
+
"model.visual.blocks.16.norm1.bias": "model-00003-of-00003.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm1.weight": "model-00003-of-00003.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm2.bias": "model-00003-of-00003.safetensors",
|
| 540 |
+
"model.visual.blocks.16.norm2.weight": "model-00003-of-00003.safetensors",
|
| 541 |
+
"model.visual.blocks.17.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 544 |
+
"model.visual.blocks.17.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 545 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 548 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 549 |
+
"model.visual.blocks.17.norm1.bias": "model-00003-of-00003.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm1.weight": "model-00003-of-00003.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm2.bias": "model-00003-of-00003.safetensors",
|
| 552 |
+
"model.visual.blocks.17.norm2.weight": "model-00003-of-00003.safetensors",
|
| 553 |
+
"model.visual.blocks.18.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 556 |
+
"model.visual.blocks.18.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 557 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 560 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 561 |
+
"model.visual.blocks.18.norm1.bias": "model-00003-of-00003.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm1.weight": "model-00003-of-00003.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm2.bias": "model-00003-of-00003.safetensors",
|
| 564 |
+
"model.visual.blocks.18.norm2.weight": "model-00003-of-00003.safetensors",
|
| 565 |
+
"model.visual.blocks.19.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 568 |
+
"model.visual.blocks.19.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 569 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 572 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 573 |
+
"model.visual.blocks.19.norm1.bias": "model-00003-of-00003.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm1.weight": "model-00003-of-00003.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm2.bias": "model-00003-of-00003.safetensors",
|
| 576 |
+
"model.visual.blocks.19.norm2.weight": "model-00003-of-00003.safetensors",
|
| 577 |
+
"model.visual.blocks.2.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 580 |
+
"model.visual.blocks.2.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 581 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 584 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 585 |
+
"model.visual.blocks.2.norm1.bias": "model-00003-of-00003.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm1.weight": "model-00003-of-00003.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm2.bias": "model-00003-of-00003.safetensors",
|
| 588 |
+
"model.visual.blocks.2.norm2.weight": "model-00003-of-00003.safetensors",
|
| 589 |
+
"model.visual.blocks.20.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 592 |
+
"model.visual.blocks.20.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 593 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 596 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 597 |
+
"model.visual.blocks.20.norm1.bias": "model-00003-of-00003.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm1.weight": "model-00003-of-00003.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm2.bias": "model-00003-of-00003.safetensors",
|
| 600 |
+
"model.visual.blocks.20.norm2.weight": "model-00003-of-00003.safetensors",
|
| 601 |
+
"model.visual.blocks.21.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 604 |
+
"model.visual.blocks.21.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 605 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 608 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 609 |
+
"model.visual.blocks.21.norm1.bias": "model-00003-of-00003.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm1.weight": "model-00003-of-00003.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm2.bias": "model-00003-of-00003.safetensors",
|
| 612 |
+
"model.visual.blocks.21.norm2.weight": "model-00003-of-00003.safetensors",
|
| 613 |
+
"model.visual.blocks.22.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 616 |
+
"model.visual.blocks.22.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 617 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 620 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 621 |
+
"model.visual.blocks.22.norm1.bias": "model-00003-of-00003.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm1.weight": "model-00003-of-00003.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm2.bias": "model-00003-of-00003.safetensors",
|
| 624 |
+
"model.visual.blocks.22.norm2.weight": "model-00003-of-00003.safetensors",
|
| 625 |
+
"model.visual.blocks.23.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 628 |
+
"model.visual.blocks.23.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 629 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 632 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 633 |
+
"model.visual.blocks.23.norm1.bias": "model-00003-of-00003.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm1.weight": "model-00003-of-00003.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm2.bias": "model-00003-of-00003.safetensors",
|
| 636 |
+
"model.visual.blocks.23.norm2.weight": "model-00003-of-00003.safetensors",
|
| 637 |
+
"model.visual.blocks.3.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 638 |
+
"model.visual.blocks.3.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 639 |
+
"model.visual.blocks.3.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 640 |
+
"model.visual.blocks.3.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 641 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 642 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 643 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 644 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 645 |
+
"model.visual.blocks.3.norm1.bias": "model-00003-of-00003.safetensors",
|
| 646 |
+
"model.visual.blocks.3.norm1.weight": "model-00003-of-00003.safetensors",
|
| 647 |
+
"model.visual.blocks.3.norm2.bias": "model-00003-of-00003.safetensors",
|
| 648 |
+
"model.visual.blocks.3.norm2.weight": "model-00003-of-00003.safetensors",
|
| 649 |
+
"model.visual.blocks.4.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 650 |
+
"model.visual.blocks.4.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 651 |
+
"model.visual.blocks.4.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 652 |
+
"model.visual.blocks.4.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 653 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 654 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 655 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 656 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 657 |
+
"model.visual.blocks.4.norm1.bias": "model-00003-of-00003.safetensors",
|
| 658 |
+
"model.visual.blocks.4.norm1.weight": "model-00003-of-00003.safetensors",
|
| 659 |
+
"model.visual.blocks.4.norm2.bias": "model-00003-of-00003.safetensors",
|
| 660 |
+
"model.visual.blocks.4.norm2.weight": "model-00003-of-00003.safetensors",
|
| 661 |
+
"model.visual.blocks.5.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 662 |
+
"model.visual.blocks.5.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 663 |
+
"model.visual.blocks.5.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 664 |
+
"model.visual.blocks.5.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 665 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 666 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 667 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 668 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 669 |
+
"model.visual.blocks.5.norm1.bias": "model-00003-of-00003.safetensors",
|
| 670 |
+
"model.visual.blocks.5.norm1.weight": "model-00003-of-00003.safetensors",
|
| 671 |
+
"model.visual.blocks.5.norm2.bias": "model-00003-of-00003.safetensors",
|
| 672 |
+
"model.visual.blocks.5.norm2.weight": "model-00003-of-00003.safetensors",
|
| 673 |
+
"model.visual.blocks.6.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 674 |
+
"model.visual.blocks.6.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 675 |
+
"model.visual.blocks.6.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 676 |
+
"model.visual.blocks.6.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 677 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 678 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 679 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 680 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 681 |
+
"model.visual.blocks.6.norm1.bias": "model-00003-of-00003.safetensors",
|
| 682 |
+
"model.visual.blocks.6.norm1.weight": "model-00003-of-00003.safetensors",
|
| 683 |
+
"model.visual.blocks.6.norm2.bias": "model-00003-of-00003.safetensors",
|
| 684 |
+
"model.visual.blocks.6.norm2.weight": "model-00003-of-00003.safetensors",
|
| 685 |
+
"model.visual.blocks.7.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 686 |
+
"model.visual.blocks.7.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 687 |
+
"model.visual.blocks.7.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 688 |
+
"model.visual.blocks.7.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 689 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 690 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 691 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 692 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 693 |
+
"model.visual.blocks.7.norm1.bias": "model-00003-of-00003.safetensors",
|
| 694 |
+
"model.visual.blocks.7.norm1.weight": "model-00003-of-00003.safetensors",
|
| 695 |
+
"model.visual.blocks.7.norm2.bias": "model-00003-of-00003.safetensors",
|
| 696 |
+
"model.visual.blocks.7.norm2.weight": "model-00003-of-00003.safetensors",
|
| 697 |
+
"model.visual.blocks.8.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 698 |
+
"model.visual.blocks.8.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 699 |
+
"model.visual.blocks.8.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 700 |
+
"model.visual.blocks.8.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 701 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 702 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 703 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 704 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 705 |
+
"model.visual.blocks.8.norm1.bias": "model-00003-of-00003.safetensors",
|
| 706 |
+
"model.visual.blocks.8.norm1.weight": "model-00003-of-00003.safetensors",
|
| 707 |
+
"model.visual.blocks.8.norm2.bias": "model-00003-of-00003.safetensors",
|
| 708 |
+
"model.visual.blocks.8.norm2.weight": "model-00003-of-00003.safetensors",
|
| 709 |
+
"model.visual.blocks.9.attn.proj.bias": "model-00003-of-00003.safetensors",
|
| 710 |
+
"model.visual.blocks.9.attn.proj.weight": "model-00003-of-00003.safetensors",
|
| 711 |
+
"model.visual.blocks.9.attn.qkv.bias": "model-00003-of-00003.safetensors",
|
| 712 |
+
"model.visual.blocks.9.attn.qkv.weight": "model-00003-of-00003.safetensors",
|
| 713 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 714 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 715 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 716 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 717 |
+
"model.visual.blocks.9.norm1.bias": "model-00003-of-00003.safetensors",
|
| 718 |
+
"model.visual.blocks.9.norm1.weight": "model-00003-of-00003.safetensors",
|
| 719 |
+
"model.visual.blocks.9.norm2.bias": "model-00003-of-00003.safetensors",
|
| 720 |
+
"model.visual.blocks.9.norm2.weight": "model-00003-of-00003.safetensors",
|
| 721 |
+
"model.visual.merger.linear_fc1.bias": "model-00003-of-00003.safetensors",
|
| 722 |
+
"model.visual.merger.linear_fc1.weight": "model-00003-of-00003.safetensors",
|
| 723 |
+
"model.visual.merger.linear_fc2.bias": "model-00003-of-00003.safetensors",
|
| 724 |
+
"model.visual.merger.linear_fc2.weight": "model-00003-of-00003.safetensors",
|
| 725 |
+
"model.visual.merger.norm.bias": "model-00003-of-00003.safetensors",
|
| 726 |
+
"model.visual.merger.norm.weight": "model-00003-of-00003.safetensors",
|
| 727 |
+
"model.visual.patch_embed.proj.bias": "model-00003-of-00003.safetensors",
|
| 728 |
+
"model.visual.patch_embed.proj.weight": "model-00003-of-00003.safetensors",
|
| 729 |
+
"model.visual.pos_embed.weight": "model-00003-of-00003.safetensors"
|
| 730 |
+
}
|
| 731 |
+
}
|
processor_config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_rescale": true,
|
| 6 |
+
"do_resize": true,
|
| 7 |
+
"image_mean": [
|
| 8 |
+
0.5,
|
| 9 |
+
0.5,
|
| 10 |
+
0.5
|
| 11 |
+
],
|
| 12 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 13 |
+
"image_std": [
|
| 14 |
+
0.5,
|
| 15 |
+
0.5,
|
| 16 |
+
0.5
|
| 17 |
+
],
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"patch_size": 16,
|
| 20 |
+
"resample": 3,
|
| 21 |
+
"rescale_factor": 0.00392156862745098,
|
| 22 |
+
"size": {
|
| 23 |
+
"longest_edge": 16777216,
|
| 24 |
+
"shortest_edge": 65536
|
| 25 |
+
},
|
| 26 |
+
"temporal_patch_size": 2
|
| 27 |
+
},
|
| 28 |
+
"processor_class": "Qwen3VLProcessor",
|
| 29 |
+
"video_processor": {
|
| 30 |
+
"do_convert_rgb": true,
|
| 31 |
+
"do_normalize": true,
|
| 32 |
+
"do_rescale": true,
|
| 33 |
+
"do_resize": true,
|
| 34 |
+
"do_sample_frames": true,
|
| 35 |
+
"fps": 2,
|
| 36 |
+
"image_mean": [
|
| 37 |
+
0.5,
|
| 38 |
+
0.5,
|
| 39 |
+
0.5
|
| 40 |
+
],
|
| 41 |
+
"image_std": [
|
| 42 |
+
0.5,
|
| 43 |
+
0.5,
|
| 44 |
+
0.5
|
| 45 |
+
],
|
| 46 |
+
"max_frames": 768,
|
| 47 |
+
"max_video_tokens": 768,
|
| 48 |
+
"merge_size": 2,
|
| 49 |
+
"min_frames": 4,
|
| 50 |
+
"patch_size": 16,
|
| 51 |
+
"resample": 3,
|
| 52 |
+
"rescale_factor": 0.00392156862745098,
|
| 53 |
+
"return_metadata": false,
|
| 54 |
+
"size": {
|
| 55 |
+
"longest_edge": 25165824,
|
| 56 |
+
"shortest_edge": 4096
|
| 57 |
+
},
|
| 58 |
+
"temporal_patch_size": 2,
|
| 59 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 60 |
+
}
|
| 61 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
|
| 3 |
+
size 19989325
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"local_files_only": false,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|
vjev.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"readout": "trailing",
|
| 3 |
+
"pause": 0,
|
| 4 |
+
"arch": "vl",
|
| 5 |
+
"max_state": 768,
|
| 6 |
+
"max_len": 2560,
|
| 7 |
+
"step": 600,
|
| 8 |
+
"lora_text_modules": 248,
|
| 9 |
+
"lora_vision_modules": 98
|
| 10 |
+
}
|
vjev_infer.py
ADDED
|
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Stand-alone inference for vjev: typed decisions from text and images, one forward pass.
|
| 2 |
+
|
| 3 |
+
pip install "transformers>=5.15" torch pillow safetensors
|
| 4 |
+
|
| 5 |
+
from vjev_infer import Vjev
|
| 6 |
+
m = Vjev("yah01/vjev-vision") # cuda, mps or cpu; bf16 on cuda, fp16 on mps
|
| 7 |
+
m.ask(state=["photo.jpg", "Frame from the warehouse camera."],
|
| 8 |
+
questions={"person": {"type": "noul", "instructions": "There is a person in this image."},
|
| 9 |
+
"where": {"type": "choice", "instructions": "Where is the forklift?",
|
| 10 |
+
"criteria": {"left": "left half", "right": "right half", "none": "no forklift"}},
|
| 11 |
+
"busy": {"type": "score", "instructions": "How cluttered is the scene?",
|
| 12 |
+
"criteria": ["empty", "sparse", "busy", "crowded"]}})
|
| 13 |
+
# -> {"person": {"type": "noul", "noul": 0.93},
|
| 14 |
+
# "where": {"type": "choice", "choice": "left", "probabilities": {...}, "confidence": ...},
|
| 15 |
+
# "busy": {"type": "score", "score": 1.4, "probabilities": {"0": ..}, "legend": {...}}}
|
| 16 |
+
|
| 17 |
+
`state` is a string, a PIL image, a path, or a list mixing them (images first is not required,
|
| 18 |
+
but text is placed after all images). Every question is scored on its own: questions never see
|
| 19 |
+
each other. Options of one choice/score question do see each other - that is the point of the
|
| 20 |
+
listwise readout - and are read in the order given.
|
| 21 |
+
|
| 22 |
+
This file is the minimum needed to use the weights; the project it comes from also has a
|
| 23 |
+
Jev-compatible HTTP server, a web console, prefix sharing across questions, and the training code.
|
| 24 |
+
"""
|
| 25 |
+
from __future__ import annotations
|
| 26 |
+
|
| 27 |
+
import json
|
| 28 |
+
from pathlib import Path
|
| 29 |
+
|
| 30 |
+
import torch
|
| 31 |
+
import torch.nn as nn
|
| 32 |
+
import torch.nn.functional as F
|
| 33 |
+
from PIL import Image
|
| 34 |
+
|
| 35 |
+
MAX_SIDE = 512 # every training image had its longer side at 512 px
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
def label(i: int) -> str:
|
| 39 |
+
s, i = "", i + 1
|
| 40 |
+
while i:
|
| 41 |
+
i, r = divmod(i - 1, 26)
|
| 42 |
+
s = chr(65 + r) + s
|
| 43 |
+
return s
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
def fit(image: Image.Image) -> Image.Image:
|
| 47 |
+
w, h = image.size
|
| 48 |
+
if max(w, h) <= MAX_SIDE:
|
| 49 |
+
return image
|
| 50 |
+
k = MAX_SIDE / max(w, h)
|
| 51 |
+
return image.resize((max(1, int(w * k)), max(1, int(h * k))))
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
class Vjev:
|
| 55 |
+
def __init__(self, repo: str, device: str | None = None, dtype=None):
|
| 56 |
+
from huggingface_hub import hf_hub_download
|
| 57 |
+
from transformers import AutoModelForImageTextToText, AutoProcessor
|
| 58 |
+
device = device or ("cuda" if torch.cuda.is_available() else "mps" if torch.backends.mps.is_available() else "cpu")
|
| 59 |
+
dtype = dtype or (torch.float16 if device == "mps" else torch.bfloat16)
|
| 60 |
+
self.device = device
|
| 61 |
+
self.proc = AutoProcessor.from_pretrained(repo)
|
| 62 |
+
self.tok = self.proc.tokenizer
|
| 63 |
+
full = AutoModelForImageTextToText.from_pretrained(repo, dtype=dtype, attn_implementation="sdpa")
|
| 64 |
+
self.trunk = full.model.to(device).eval() # the vocabulary projection is unused
|
| 65 |
+
self.trunk.config.use_cache = False
|
| 66 |
+
self.cfg = full.config
|
| 67 |
+
get = (lambda f: Path(repo) / f) if Path(repo).is_dir() else (lambda f: hf_hub_download(repo, f))
|
| 68 |
+
self.meta = json.loads(Path(get("vjev.json")).read_text())
|
| 69 |
+
hidden = self.cfg.text_config.hidden_size
|
| 70 |
+
self.head = nn.Linear(hidden, 1, dtype=torch.float32)
|
| 71 |
+
self.head.load_state_dict(torch.load(get("head.pt"), map_location="cpu"))
|
| 72 |
+
self.head.to(device)
|
| 73 |
+
self._linear_patch_embed()
|
| 74 |
+
|
| 75 |
+
def _linear_patch_embed(self):
|
| 76 |
+
# The ViT's patch embedding is a Conv3d whose kernel == stride == input: a linear map of
|
| 77 |
+
# each flattened patch. As a Conv3d torch runs it very slowly in bf16 on CUDA (measured
|
| 78 |
+
# 35 s per 8 images); the same weights as one matmul take milliseconds.
|
| 79 |
+
pe = self.trunk.visual.patch_embed
|
| 80 |
+
|
| 81 |
+
def forward(x):
|
| 82 |
+
w = pe.proj.weight
|
| 83 |
+
return F.linear(x.to(w.dtype).view(-1, w[0].numel()), w.view(w.size(0), -1), pe.proj.bias)
|
| 84 |
+
pe.forward = forward
|
| 85 |
+
|
| 86 |
+
# ---- rendering -------------------------------------------------------------
|
| 87 |
+
def _ids(self, s: str) -> list[int]:
|
| 88 |
+
return self.tok(s, add_special_tokens=False).input_ids
|
| 89 |
+
|
| 90 |
+
def _render(self, images: list[Image.Image], grids: list[int], text: str, q: dict):
|
| 91 |
+
cfg = self.cfg
|
| 92 |
+
ids, types = [], []
|
| 93 |
+
for i, n in enumerate(grids):
|
| 94 |
+
if len(grids) > 1: # several images are numbered
|
| 95 |
+
lab = self._ids(("" if i == 0 else "\n") + f"Picture {i + 1}: ")
|
| 96 |
+
ids += lab; types += [0] * len(lab)
|
| 97 |
+
ids += [cfg.vision_start_token_id] + [cfg.image_token_id] * n + [cfg.vision_end_token_id]
|
| 98 |
+
types += [0] + [1] * n + [0]
|
| 99 |
+
if text:
|
| 100 |
+
ids += self._ids(text)
|
| 101 |
+
ids += self._ids(f"\n\nQuestion: {q['instructions']}")
|
| 102 |
+
t = q["type"]
|
| 103 |
+
opts = [] if t == "noul" else (list(q["criteria"].values()) if isinstance(q["criteria"], dict) else list(q["criteria"]))
|
| 104 |
+
slots = []
|
| 105 |
+
if t == "noul":
|
| 106 |
+
slots.append(len(ids) - 1)
|
| 107 |
+
else:
|
| 108 |
+
for i, o in enumerate(opts):
|
| 109 |
+
ids += self._ids(f"\n({label(i)}) {o}")
|
| 110 |
+
ids += self._ids("\nAnswer:") # trailing readout: every slot after the whole list
|
| 111 |
+
for i in range(len(opts)):
|
| 112 |
+
ids += self._ids(f" ({label(i)})")
|
| 113 |
+
slots.append(len(ids) - 1)
|
| 114 |
+
types += [0] * (len(ids) - len(types))
|
| 115 |
+
if len(ids) > self.meta.get("max_len", 2560) + 1536:
|
| 116 |
+
raise ValueError(f"question {q['instructions'][:40]!r}: {len(ids)} tokens is over the limit")
|
| 117 |
+
return ids, types, slots
|
| 118 |
+
|
| 119 |
+
# ---- scoring ---------------------------------------------------------------
|
| 120 |
+
@torch.no_grad()
|
| 121 |
+
def ask(self, state, questions: dict, batch_rows: int = 8) -> dict:
|
| 122 |
+
items = state if isinstance(state, list) else [state]
|
| 123 |
+
images, texts = [], []
|
| 124 |
+
for x in items:
|
| 125 |
+
if isinstance(x, Image.Image):
|
| 126 |
+
images.append(fit(x.convert("RGB")))
|
| 127 |
+
elif isinstance(x, (str, Path)) and Path(x).is_file():
|
| 128 |
+
images.append(fit(Image.open(x).convert("RGB")))
|
| 129 |
+
else:
|
| 130 |
+
texts.append(str(x))
|
| 131 |
+
text = "\n".join(texts)
|
| 132 |
+
mm, grids = {}, []
|
| 133 |
+
if images:
|
| 134 |
+
out = self.proc.image_processor(images=images, return_tensors="pt")
|
| 135 |
+
merge = self.proc.image_processor.merge_size ** 2
|
| 136 |
+
grids = [int(g.prod()) // merge for g in out["image_grid_thw"]]
|
| 137 |
+
mm = {"pixel_values": out["pixel_values"].to(self.device), "image_grid_thw": out["image_grid_thw"].to(self.device)}
|
| 138 |
+
rendered = {name: self._render(images, grids, text, q) for name, q in questions.items()}
|
| 139 |
+
pad = self.tok.pad_token_id if self.tok.pad_token_id is not None else self.tok.eos_token_id
|
| 140 |
+
answers, names = {}, list(rendered)
|
| 141 |
+
for b0 in range(0, len(names), batch_rows):
|
| 142 |
+
batch = names[b0:b0 + batch_rows]
|
| 143 |
+
S = max(len(rendered[n][0]) for n in batch)
|
| 144 |
+
ids = torch.full((len(batch), S), pad, dtype=torch.long)
|
| 145 |
+
att = torch.zeros((len(batch), S), dtype=torch.long)
|
| 146 |
+
typ = torch.zeros((len(batch), S), dtype=torch.long)
|
| 147 |
+
for i, n in enumerate(batch): # left-padded, as in training
|
| 148 |
+
r_ids, r_typ, _ = rendered[n]
|
| 149 |
+
ids[i, S - len(r_ids):] = torch.tensor(r_ids); att[i, S - len(r_ids):] = 1
|
| 150 |
+
typ[i, S - len(r_ids):] = torch.tensor(r_typ)
|
| 151 |
+
kw = dict(mm)
|
| 152 |
+
if images:
|
| 153 |
+
kw["pixel_values"] = mm["pixel_values"].repeat(len(batch), 1)
|
| 154 |
+
kw["image_grid_thw"] = mm["image_grid_thw"].repeat(len(batch), 1)
|
| 155 |
+
kw["mm_token_type_ids"] = typ.to(self.device)
|
| 156 |
+
h = self.trunk(input_ids=ids.to(self.device), attention_mask=att.to(self.device), **kw).last_hidden_state
|
| 157 |
+
for i, n in enumerate(batch):
|
| 158 |
+
slots = torch.tensor(rendered[n][2], device=self.device) + (S - len(rendered[n][0]))
|
| 159 |
+
logits = self.head(h[i, slots].float()).squeeze(-1)
|
| 160 |
+
answers[n] = self._answer(questions[n], logits)
|
| 161 |
+
return answers
|
| 162 |
+
|
| 163 |
+
@staticmethod
|
| 164 |
+
def _answer(q: dict, logits: torch.Tensor) -> dict:
|
| 165 |
+
t = q["type"]
|
| 166 |
+
if t == "noul":
|
| 167 |
+
return {"type": "noul", "noul": round(float(torch.sigmoid(logits[0])), 4)}
|
| 168 |
+
p = torch.softmax(logits, -1).tolist()
|
| 169 |
+
n = len(p)
|
| 170 |
+
conf = round(max(0.0, (max(p) - 1 / n) / (1 - 1 / n)), 4) if n > 1 else 0.0
|
| 171 |
+
if t == "choice":
|
| 172 |
+
keys = list(q["criteria"])
|
| 173 |
+
probs = {k: round(v, 4) for k, v in zip(keys, p)}
|
| 174 |
+
return {"type": "choice", "choice": max(probs, key=probs.get), "confidence": conf, "probabilities": probs}
|
| 175 |
+
levels = list(q["criteria"])
|
| 176 |
+
return {"type": "score", "score": round(sum(i * v for i, v in enumerate(p)), 4), "confidence": conf,
|
| 177 |
+
"legend": {str(i): l for i, l in enumerate(levels)},
|
| 178 |
+
"probabilities": {str(i): round(v, 4) for i, v in enumerate(p)}}
|