Upload folder using huggingface_hub
Browse files- README.md +47 -0
- comparison_graph.png +0 -0
- config.json +34 -0
- generation_config.json +4 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +4 -0
- tokenizer.json +0 -0
- tokenizer_config.json +18 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- pruned
|
| 5 |
+
- python
|
| 6 |
+
- optimized
|
| 7 |
+
base_model: openai-community/openai-gpt
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# openai-gpt-python-medium
|
| 11 |
+
|
| 12 |
+
This model is a **medium** pruned version of [openai-community/openai-gpt](https://huggingface.co/openai-community/openai-gpt), specialized for **PYTHON** tasks.
|
| 13 |
+
|
| 14 |
+
## Pruning Details
|
| 15 |
+
|
| 16 |
+
- **Base Model**: openai-community/openai-gpt
|
| 17 |
+
- **Specialization**: Python
|
| 18 |
+
- **Prune Mode**: Medium
|
| 19 |
+
- **Method**: Activation-based weight pruning
|
| 20 |
+
|
| 21 |
+
## Performance Comparison
|
| 22 |
+
|
| 23 |
+
| Category | Original | Pruned |
|
| 24 |
+
|----------|----------|--------|
|
| 25 |
+
| Python | 0.0% | 0.0% |
|
| 26 |
+
| HTML | 0.0% | 0.0% |
|
| 27 |
+
| Trivia | 0.0% | 0.0% |
|
| 28 |
+
| Math | 0.0% | 0.0% |
|
| 29 |
+
| Reasoning | 0.0% | 0.0% |
|
| 30 |
+
| Medical | 0.0% | 0.0% |
|
| 31 |
+
| Linux | 0.0% | 0.0% |
|
| 32 |
+
| Writing | 0.0% | 0.0% |
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
|
| 36 |
+
## Usage
|
| 37 |
+
|
| 38 |
+
```python
|
| 39 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 40 |
+
|
| 41 |
+
model = AutoModelForCausalLM.from_pretrained("CompactAI/openai-gpt-python-medium")
|
| 42 |
+
tokenizer = AutoTokenizer.from_pretrained("CompactAI/openai-gpt-python-medium")
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
## License
|
| 46 |
+
|
| 47 |
+
This model inherits the license from the base model.
|
comparison_graph.png
ADDED
|
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"afn": "gelu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"OpenAIGPTLMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"dtype": "float16",
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"initializer_range": 0.02,
|
| 10 |
+
"layer_norm_epsilon": 1e-05,
|
| 11 |
+
"model_type": "openai-gpt",
|
| 12 |
+
"n_ctx": 512,
|
| 13 |
+
"n_embd": 768,
|
| 14 |
+
"n_head": 12,
|
| 15 |
+
"n_layer": 12,
|
| 16 |
+
"n_positions": 512,
|
| 17 |
+
"n_special": 0,
|
| 18 |
+
"predict_special_tokens": true,
|
| 19 |
+
"resid_pdrop": 0.1,
|
| 20 |
+
"summary_activation": null,
|
| 21 |
+
"summary_first_dropout": 0.1,
|
| 22 |
+
"summary_proj_to_labels": true,
|
| 23 |
+
"summary_type": "cls_index",
|
| 24 |
+
"summary_use_proj": true,
|
| 25 |
+
"task_specific_params": {
|
| 26 |
+
"text-generation": {
|
| 27 |
+
"do_sample": true,
|
| 28 |
+
"max_length": 50
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"transformers_version": "4.57.6",
|
| 32 |
+
"use_cache": false,
|
| 33 |
+
"vocab_size": 40478
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.57.6"
|
| 4 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e90e9b5960d2ecabf596f5f9ed320382eb4004983e69ec341d6c17cc52039c8
|
| 3 |
+
size 233084296
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": "<unk>",
|
| 3 |
+
"unk_token": "<unk>"
|
| 4 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
}
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": false,
|
| 13 |
+
"extra_special_tokens": {},
|
| 14 |
+
"model_max_length": 512,
|
| 15 |
+
"pad_token": "<unk>",
|
| 16 |
+
"tokenizer_class": "OpenAIGPTTokenizer",
|
| 17 |
+
"unk_token": "<unk>"
|
| 18 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|