alicegoesdown's picture
Training in progress, step 150
3abc8ea verified
{
"alpha_pattern": {},
"auto_mapping": null,
"base_model_name_or_path": "facebook/opt-1.3b",
"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": 128,
"lora_dropout": 0.2,
"megatron_config": null,
"megatron_core": "megatron.core",
"modules_to_save": null,
"peft_type": "LORA",
"r": 64,
"rank_pattern": {},
"revision": null,
"target_modules": [
"model.decoder.layers.12.self_attn.out_proj",
"model.decoder.layers.22.self_attn.v_proj",
"model.decoder.layers.20.fc2",
"model.decoder.layers.2.fc2",
"model.decoder.layers.19.self_attn.v_proj",
"model.decoder.layers.4.fc1",
"model.decoder.layers.16.fc2",
"model.decoder.layers.5.self_attn.k_proj",
"model.decoder.layers.23.self_attn.v_proj",
"model.decoder.layers.1.self_attn.q_proj",
"model.decoder.layers.3.fc1",
"model.decoder.layers.21.self_attn.q_proj",
"model.decoder.layers.16.self_attn.q_proj",
"model.decoder.layers.19.fc2",
"model.decoder.layers.2.self_attn.v_proj",
"model.decoder.embed_tokens",
"model.decoder.layers.23.self_attn.out_proj",
"model.decoder.layers.3.fc2",
"model.decoder.layers.8.self_attn.v_proj",
"model.decoder.layers.13.fc1",
"model.decoder.layers.2.self_attn.k_proj",
"model.decoder.layers.4.self_attn.out_proj",
"model.decoder.layers.1.fc2",
"model.decoder.layers.14.fc2",
"model.decoder.layers.7.fc1",
"model.decoder.layers.17.fc1",
"model.decoder.layers.10.self_attn.q_proj",
"model.decoder.layers.22.self_attn.k_proj",
"model.decoder.layers.23.fc1",
"model.decoder.layers.9.self_attn.v_proj",
"model.decoder.layers.22.self_attn.q_proj",
"model.decoder.layers.23.fc2",
"model.decoder.embed_positions",
"model.decoder.layers.0.fc2",
"model.decoder.layers.0.fc1",
"model.decoder.layers.4.self_attn.k_proj",
"model.decoder.layers.20.self_attn.q_proj",
"model.decoder.layers.18.self_attn.k_proj",
"model.decoder.layers.5.fc2",
"model.decoder.layers.18.self_attn.q_proj",
"model.decoder.layers.12.self_attn.k_proj",
"model.decoder.layers.11.self_attn.k_proj",
"model.decoder.layers.1.self_attn.v_proj",
"model.decoder.layers.16.fc1",
"model.decoder.layers.6.self_attn.k_proj",
"model.decoder.layers.15.self_attn.out_proj",
"model.decoder.layers.23.self_attn.k_proj",
"model.decoder.layers.21.self_attn.v_proj",
"model.decoder.layers.9.self_attn.out_proj",
"model.decoder.layers.8.self_attn.out_proj",
"model.decoder.layers.19.self_attn.k_proj",
"model.decoder.layers.20.self_attn.out_proj",
"model.decoder.layers.7.self_attn.v_proj",
"model.decoder.layers.6.self_attn.v_proj",
"model.decoder.layers.22.fc2",
"model.decoder.layers.11.self_attn.q_proj",
"model.decoder.layers.13.fc2",
"model.decoder.layers.10.fc2",
"model.decoder.layers.8.self_attn.q_proj",
"model.decoder.layers.15.self_attn.k_proj",
"model.decoder.layers.0.self_attn.k_proj",
"model.decoder.layers.17.self_attn.k_proj",
"model.decoder.layers.3.self_attn.k_proj",
"model.decoder.layers.13.self_attn.q_proj",
"model.decoder.layers.9.self_attn.k_proj",
"model.decoder.layers.14.self_attn.k_proj",
"model.decoder.layers.23.self_attn.q_proj",
"model.decoder.layers.15.self_attn.q_proj",
"model.decoder.layers.22.fc1",
"model.decoder.layers.18.self_attn.v_proj",
"model.decoder.layers.12.fc2",
"model.decoder.layers.5.self_attn.v_proj",
"model.decoder.layers.7.self_attn.out_proj",
"model.decoder.layers.21.fc2",
"model.decoder.layers.6.fc1",
"model.decoder.layers.10.self_attn.v_proj",
"model.decoder.layers.22.self_attn.out_proj",
"lm_head",
"model.decoder.layers.14.self_attn.out_proj",
"model.decoder.layers.17.self_attn.v_proj",
"model.decoder.layers.11.self_attn.v_proj"
],
"task_type": "CAUSAL_LM",
"use_dora": false,
"use_rslora": false
}