| { | |
| "architectures": [ | |
| "JudgeModel", | |
| "JudgeCasualLMHead", | |
| "JudgeWithQA", | |
| "JudgeClassifier" | |
| ], | |
| "inference_mode": true, | |
| "batch_size": 32, | |
| "bias": true, | |
| "block_size": 512, | |
| "dim_feedforward": 3072, | |
| "dropout": 0.1, | |
| "ff_expansion_factor": 4, | |
| "hidden_act": "gelu", | |
| "hidden_size": 768, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1", | |
| "2": "LABEL_2", | |
| "3": "LABEL_3", | |
| "4": "LABEL_4" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1, | |
| "LABEL_2": 2, | |
| "LABEL_3": 3, | |
| "LABEL_4": 4 | |
| }, | |
| "label_smoothing": 0.1, | |
| "learning_rate": 0.0003, | |
| "log_interval": 100, | |
| "max_grad_norm": 1.0, | |
| "max_position_embeddings": 512, | |
| "model_type": "gpt2", | |
| "n_embd": 768, | |
| "n_head": 12, | |
| "n_layer": 12, | |
| "output_dir": "C:/Users/wonde/output", | |
| "torch_dtype": "float32", | |
| "total_steps": 10000, | |
| "transformers_version": "4.43.3", | |
| "vocab_size": 50257, | |
| "warmup_steps": 1000, | |
| "weight_decay": 0.01, | |
| "task_heads": { | |
| "lm_head": { | |
| "type": "JudgeCasualLMHead", | |
| "params": { | |
| "vocab_size": 50257 | |
| } | |
| }, | |
| "qa_head": { | |
| "type": "JudgeWithQA", | |
| "params": { | |
| "num_labels": 2 | |
| } | |
| }, | |
| "classifier_head": { | |
| "type": "JudgeClassifier", | |
| "params": { | |
| "num_labels": 5 | |
| } | |
| } | |
| }, | |
| "task_specific_params": { | |
| "text-generation": { | |
| "do_sample": true, | |
| "max_length": 100 | |
| }, | |
| "question-answering": { | |
| "max_answer_length": 30 | |
| }, | |
| "sequence-classification": { | |
| "eval_steps": 500 | |
| } | |
| } | |
| } |