| { | |
| "alpha_pattern": {}, | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "inception-mbzuai/jais-13b-chat", | |
| "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": 64, | |
| "lora_dropout": 0.1, | |
| "megatron_config": null, | |
| "megatron_core": "megatron.core", | |
| "modules_to_save": null, | |
| "peft_type": "LORA", | |
| "r": 16, | |
| "rank_pattern": {}, | |
| "revision": null, | |
| "target_modules": { | |
| "lm_head": "Linear4bit", | |
| "transformer.h.32.attn.c_attn": "Linear4bit", | |
| "transformer.h.32.attn.c_proj": "Linear4bit", | |
| "transformer.h.32.mlp.c_fc": "Linear4bit", | |
| "transformer.h.32.mlp.c_fc2": "Linear4bit", | |
| "transformer.h.32.mlp.c_proj": "Linear4bit", | |
| "transformer.h.33.attn.c_attn": "Linear4bit", | |
| "transformer.h.33.attn.c_proj": "Linear4bit", | |
| "transformer.h.33.mlp.c_fc": "Linear4bit", | |
| "transformer.h.33.mlp.c_fc2": "Linear4bit", | |
| "transformer.h.33.mlp.c_proj": "Linear4bit", | |
| "transformer.h.35.attn.c_attn": "Linear4bit", | |
| "transformer.h.35.attn.c_proj": "Linear4bit", | |
| "transformer.h.35.mlp.c_fc": "Linear4bit", | |
| "transformer.h.35.mlp.c_fc2": "Linear4bit", | |
| "transformer.h.35.mlp.c_proj": "Linear4bit", | |
| "transformer.h.39.attn.c_attn": "Linear4bit", | |
| "transformer.h.39.attn.c_proj": "Linear4bit", | |
| "transformer.h.39.mlp.c_fc": "Linear4bit", | |
| "transformer.h.39.mlp.c_fc2": "Linear4bit", | |
| "transformer.h.39.mlp.c_proj": "Linear4bit" | |
| }, | |
| "task_type": "CAUSAL_LM", | |
| "use_dora": false, | |
| "use_rslora": false | |
| } |