Instructions to use GreenGenomicsLab/mamba-tinytrain1.4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use GreenGenomicsLab/mamba-tinytrain1.4b with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-1.4b-hf") model = PeftModel.from_pretrained(base_model, "GreenGenomicsLab/mamba-tinytrain1.4b") - Notebooks
- Google Colab
- Kaggle
File size: 685 Bytes
4cf1e5f | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | {
"alpha_pattern": {},
"auto_mapping": null,
"base_model_name_or_path": "state-spaces/mamba-1.4b-hf",
"bias": "none",
"fan_in_fan_out": false,
"inference_mode": true,
"init_lora_weights": true,
"layer_replication": null,
"layers_pattern": null,
"layers_to_transform": null,
"loftq_config": {},
"lora_alpha": 8,
"lora_dropout": 0.0,
"megatron_config": null,
"megatron_core": "megatron.core",
"modules_to_save": null,
"peft_type": "LORA",
"r": 8,
"rank_pattern": {},
"revision": null,
"target_modules": [
"embeddings",
"x_proj",
"in_proj",
"out_proj"
],
"task_type": "CAUSAL_LM",
"use_dora": false,
"use_rslora": false
}
|