weed / adapter_config.json
AdamStormed22's picture
Training in progress, step 20
0438906 verified
{
"alpha_pattern": {},
"auto_mapping": {
"base_model_class": "LlavaNextForConditionalGeneration",
"parent_library": "transformers.models.llava_next.modeling_llava_next"
},
"base_model_name_or_path": "ibm-granite/granite-vision-3.1-2b-preview",
"bias": "none",
"corda_config": null,
"eva_config": null,
"exclude_modules": null,
"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": 8,
"lora_bias": false,
"lora_dropout": 0.1,
"megatron_config": null,
"megatron_core": "megatron.core",
"modules_to_save": null,
"peft_type": "LORA",
"r": 8,
"rank_pattern": {},
"revision": null,
"target_modules": [
"model.layers.20.self_attn.v_proj",
"30.self_attn.q_proj",
"o_proj",
"model.layers.3.self_attn.q_proj",
"gate_proj",
"up_proj",
"model.layers.4.self_attn.k_proj",
"model.layers.6.self_attn.q_proj",
"model.layers.15.self_attn.k_proj",
"model.layers.15.self_attn.v_proj",
"model.layers.20.self_attn.q_proj",
"28.self_attn.k_proj",
"model.layers.7.self_attn.k_proj",
"model.layers.24.self_attn.v_proj",
"28.self_attn.v_proj",
"37.self_attn.k_proj",
"27.self_attn.q_proj",
"model.layers.7.self_attn.q_proj",
"36.self_attn.v_proj",
"model.layers.11.self_attn.v_proj",
"model.layers.13.self_attn.q_proj",
"model.layers.24.self_attn.q_proj",
"model.layers.4.self_attn.q_proj",
"model.layers.23.self_attn.v_proj",
"33.self_attn.v_proj",
"33.self_attn.k_proj",
"model.layers.2.self_attn.v_proj",
"model.layers.26.self_attn.k_proj",
"model.layers.12.self_attn.v_proj",
"model.layers.17.self_attn.v_proj",
"model.layers.14.self_attn.k_proj",
"model.layers.22.self_attn.v_proj",
"model.layers.16.self_attn.q_proj",
"model.layers.13.self_attn.v_proj",
"model.layers.13.self_attn.k_proj",
"model.layers.21.self_attn.q_proj",
"38.self_attn.v_proj",
"27.self_attn.k_proj",
"27.self_attn.v_proj",
"model.layers.3.self_attn.k_proj",
"model.layers.21.self_attn.k_proj",
"model.layers.19.self_attn.k_proj",
"38.self_attn.q_proj",
"model.layers.25.self_attn.q_proj",
"model.layers.4.self_attn.v_proj",
"30.self_attn.k_proj",
"model.layers.5.self_attn.v_proj",
"model.layers.18.self_attn.v_proj",
"down_proj",
"34.self_attn.v_proj",
"model.layers.11.self_attn.q_proj",
"32.self_attn.q_proj",
"29.self_attn.q_proj",
"model.layers.17.self_attn.k_proj",
"31.self_attn.q_proj",
"model.layers.3.self_attn.v_proj",
"model.layers.1.self_attn.q_proj",
"model.layers.0.self_attn.k_proj",
"model.layers.2.self_attn.k_proj",
"model.layers.24.self_attn.k_proj",
"model.layers.23.self_attn.q_proj",
"37.self_attn.v_proj",
"model.layers.6.self_attn.k_proj",
"model.layers.0.self_attn.v_proj",
"model.layers.15.self_attn.q_proj",
"38.self_attn.k_proj",
"33.self_attn.q_proj",
"model.layers.6.self_attn.v_proj",
"model.layers.22.self_attn.q_proj",
"model.layers.9.self_attn.q_proj",
"model.layers.17.self_attn.q_proj",
"29.self_attn.k_proj",
"model.layers.26.self_attn.q_proj",
"model.layers.21.self_attn.v_proj",
"model.layers.18.self_attn.k_proj",
"model.layers.14.self_attn.v_proj",
"model.layers.8.self_attn.v_proj",
"model.layers.22.self_attn.k_proj",
"39.self_attn.k_proj",
"model.layers.16.self_attn.k_proj",
"model.layers.12.self_attn.k_proj",
"model.layers.1.self_attn.v_proj",
"model.layers.7.self_attn.v_proj",
"35.self_attn.k_proj",
"37.self_attn.q_proj",
"31.self_attn.k_proj",
"39.self_attn.q_proj",
"35.self_attn.v_proj",
"model.layers.16.self_attn.v_proj",
"model.layers.26.self_attn.v_proj",
"32.self_attn.v_proj",
"model.layers.10.self_attn.q_proj",
"model.layers.5.self_attn.q_proj",
"36.self_attn.k_proj",
"36.self_attn.q_proj",
"model.layers.11.self_attn.k_proj",
"model.layers.18.self_attn.q_proj",
"model.layers.8.self_attn.q_proj",
"model.layers.23.self_attn.k_proj",
"34.self_attn.k_proj",
"34.self_attn.q_proj",
"model.layers.1.self_attn.k_proj",
"model.layers.10.self_attn.k_proj",
"model.layers.9.self_attn.k_proj",
"model.layers.12.self_attn.q_proj",
"model.layers.25.self_attn.v_proj",
"model.layers.8.self_attn.k_proj",
"model.layers.10.self_attn.v_proj",
"model.layers.2.self_attn.q_proj",
"35.self_attn.q_proj",
"model.layers.9.self_attn.v_proj",
"30.self_attn.v_proj",
"model.layers.14.self_attn.q_proj",
"31.self_attn.v_proj",
"model.layers.0.self_attn.q_proj",
"model.layers.19.self_attn.v_proj",
"model.layers.20.self_attn.k_proj",
"28.self_attn.q_proj",
"29.self_attn.v_proj",
"model.layers.25.self_attn.k_proj",
"model.layers.19.self_attn.q_proj",
"32.self_attn.k_proj",
"39.self_attn.v_proj",
"model.layers.5.self_attn.k_proj"
],
"task_type": null,
"trainable_token_indices": null,
"use_dora": true,
"use_rslora": false
}