| { | |
| "alpha_pattern": {}, | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "distilbert-base-uncased", | |
| "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": 32, | |
| "lora_dropout": 0.1, | |
| "megatron_config": null, | |
| "megatron_core": "megatron.core", | |
| "modules_to_save": [ | |
| "classifier", | |
| "score" | |
| ], | |
| "peft_type": "LORA", | |
| "r": 16, | |
| "rank_pattern": {}, | |
| "revision": null, | |
| "target_modules": [ | |
| "distilbert.transformer.layer.0.attention.k_lin", | |
| "distilbert.transformer.layer.1.ffn.lin2", | |
| "distilbert.transformer.layer.0.attention.out_lin", | |
| "distilbert.transformer.layer.1.attention.v_lin", | |
| "distilbert.transformer.layer.2.attention.k_lin", | |
| "distilbert.transformer.layer.2.ffn.lin2", | |
| "distilbert.transformer.layer.0.attention.v_lin", | |
| "distilbert.transformer.layer.0.ffn.lin2", | |
| "distilbert.transformer.layer.2.attention.out_lin", | |
| "distilbert.transformer.layer.0.ffn.lin1", | |
| "distilbert.transformer.layer.4.attention.k_lin", | |
| "distilbert.transformer.layer.1.attention.q_lin", | |
| "distilbert.transformer.layer.2.attention.v_lin", | |
| "distilbert.transformer.layer.1.ffn.lin1", | |
| "distilbert.transformer.layer.3.ffn.lin2", | |
| "distilbert.transformer.layer.4.attention.v_lin", | |
| "distilbert.transformer.layer.4.attention.q_lin", | |
| "distilbert.transformer.layer.3.attention.v_lin", | |
| "distilbert.transformer.layer.3.ffn.lin1", | |
| "distilbert.transformer.layer.0.attention.q_lin", | |
| "distilbert.transformer.layer.4.attention.out_lin", | |
| "distilbert.transformer.layer.5.attention.k_lin", | |
| "distilbert.transformer.layer.1.attention.k_lin", | |
| "distilbert.transformer.layer.3.attention.out_lin", | |
| "distilbert.transformer.layer.3.attention.k_lin", | |
| "distilbert.transformer.layer.5.attention.q_lin", | |
| "distilbert.transformer.layer.5.ffn.lin1", | |
| "distilbert.transformer.layer.5.attention.v_lin", | |
| "distilbert.transformer.layer.4.ffn.lin2", | |
| "distilbert.transformer.layer.5.ffn.lin2", | |
| "distilbert.transformer.layer.2.ffn.lin1", | |
| "distilbert.transformer.layer.5.attention.out_lin", | |
| "distilbert.transformer.layer.1.attention.out_lin", | |
| "distilbert.transformer.layer.3.attention.q_lin", | |
| "distilbert.transformer.layer.2.attention.q_lin", | |
| "distilbert.transformer.layer.4.ffn.lin1" | |
| ], | |
| "task_type": "SEQ_CLS", | |
| "use_dora": false, | |
| "use_rslora": false | |
| } |