|
{ |
|
"alpha_pattern": {}, |
|
"auto_mapping": { |
|
"base_model_class": "MllamaForConditionalGeneration", |
|
"parent_library": "transformers.models.mllama.modeling_mllama" |
|
}, |
|
"base_model_name_or_path": "looppayments/llama-11b-vision-instruct", |
|
"bias": "none", |
|
"fan_in_fan_out": false, |
|
"inference_mode": true, |
|
"init_lora_weights": "gaussian", |
|
"layer_replication": null, |
|
"layers_pattern": null, |
|
"layers_to_transform": null, |
|
"loftq_config": {}, |
|
"lora_alpha": 128, |
|
"lora_dropout": 0.1, |
|
"megatron_config": null, |
|
"megatron_core": "megatron.core", |
|
"modules_to_save": null, |
|
"peft_type": "LORA", |
|
"r": 128, |
|
"rank_pattern": {}, |
|
"revision": null, |
|
"target_modules": [ |
|
"model.layers.25.self_attn.q_proj", |
|
"38.self_attn.v_proj", |
|
"cross_attn.q_proj", |
|
"34.self_attn.q_proj", |
|
"model.layers.25.self_attn.v_proj", |
|
"39.self_attn.v_proj", |
|
"32.self_attn.q_proj", |
|
"33.self_attn.v_proj", |
|
"37.self_attn.q_proj", |
|
"model.layers.30.self_attn.q_proj", |
|
"33.self_attn.q_proj", |
|
"model.layers.24.self_attn.q_proj", |
|
"model.layers.27.self_attn.v_proj", |
|
"model.layers.28.self_attn.v_proj", |
|
"36.self_attn.v_proj", |
|
"38.self_attn.q_proj", |
|
"35.self_attn.q_proj", |
|
"cross_attn.v_proj", |
|
"model.layers.30.self_attn.v_proj", |
|
"model.layers.27.self_attn.q_proj", |
|
"39.self_attn.q_proj", |
|
"36.self_attn.q_proj", |
|
"model.layers.31.self_attn.v_proj", |
|
"model.layers.24.self_attn.v_proj", |
|
"model.layers.28.self_attn.q_proj", |
|
"model.layers.29.self_attn.q_proj", |
|
"model.layers.26.self_attn.v_proj", |
|
"35.self_attn.v_proj", |
|
"model.layers.31.self_attn.q_proj", |
|
"model.layers.26.self_attn.q_proj", |
|
"37.self_attn.v_proj", |
|
"32.self_attn.v_proj", |
|
"34.self_attn.v_proj", |
|
"model.layers.29.self_attn.v_proj" |
|
], |
|
"task_type": null, |
|
"use_dora": false, |
|
"use_rslora": false |
|
} |