| [2025-06-30 10:56:18,436][__main__][INFO] - cache_dir: /tmp/ | |
| dataset: | |
| name: kamel-usp/aes_enem_dataset | |
| split: JBCS2025 | |
| training_params: | |
| seed: 42 | |
| num_train_epochs: 20 | |
| logging_steps: 100 | |
| metric_for_best_model: QWK | |
| bf16: true | |
| bootstrap: | |
| enabled: true | |
| n_bootstrap: 10000 | |
| bootstrap_seed: 42 | |
| metrics: | |
| - QWK | |
| - Macro_F1 | |
| - Weighted_F1 | |
| post_training_results: | |
| model_path: /workspace/jbcs2025/outputs/2025-03-24/20-42-59 | |
| experiments: | |
| model: | |
| name: microsoft/phi-4 | |
| type: phi4_classification_lora | |
| num_labels: 6 | |
| output_dir: ./results/ | |
| logging_dir: ./logs/ | |
| best_model_dir: ./results/best_model | |
| lora_r: 8 | |
| lora_dropout: 0.05 | |
| lora_alpha: 16 | |
| lora_target_modules: all-linear | |
| checkpoint_path: '' | |
| tokenizer: | |
| name: microsoft/phi-4 | |
| dataset: | |
| grade_index: 3 | |
| use_full_context: true | |
| training_params: | |
| weight_decay: 0.01 | |
| warmup_ratio: 0.1 | |
| learning_rate: 5.0e-05 | |
| train_batch_size: 1 | |
| eval_batch_size: 16 | |
| gradient_accumulation_steps: 16 | |
| gradient_checkpointing: false | |
| [2025-06-30 10:56:22,440][__main__][INFO] - GPU 0: NVIDIA H200 | TDP ≈ 700 W | |
| [2025-06-30 10:56:22,440][__main__][INFO] - Starting the Fine Tuning training process. | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file vocab.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/vocab.json | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file merges.txt from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/merges.txt | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file tokenizer.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer.json | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file added_tokens.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/added_tokens.json | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file special_tokens_map.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/special_tokens_map.json | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file tokenizer_config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer_config.json | |
| [2025-06-30 10:56:28,113][transformers.tokenization_utils_base][INFO] - loading file chat_template.jinja from cache at None | |
| [2025-06-30 10:56:28,258][__main__][INFO] - Tokenizer function parameters- Padding:longest; Truncation: False; Use Full Context: True | |
| [2025-06-30 10:56:30,029][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 10:56:30,030][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1", | |
| "2": "LABEL_2", | |
| "3": "LABEL_3", | |
| "4": "LABEL_4", | |
| "5": "LABEL_5" | |
| }, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1, | |
| "LABEL_2": 2, | |
| "LABEL_3": 3, | |
| "LABEL_4": 4, | |
| "LABEL_5": 5 | |
| }, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 10:56:30,169][transformers.modeling_utils][INFO] - loading weights file model.safetensors from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/model.safetensors.index.json | |
| [2025-06-30 10:56:30,170][transformers.modeling_utils][INFO] - Will use torch_dtype=torch.bfloat16 as defined in model's config object | |
| [2025-06-30 10:56:30,170][transformers.modeling_utils][INFO] - Instantiating Phi3ForSequenceClassification model under default dtype torch.bfloat16. | |
| [2025-06-30 10:56:37,015][transformers.modeling_utils][INFO] - Some weights of the model checkpoint at microsoft/phi-4 were not used when initializing Phi3ForSequenceClassification: ['lm_head.weight'] | |
| - This IS expected if you are initializing Phi3ForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). | |
| - This IS NOT expected if you are initializing Phi3ForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). | |
| [2025-06-30 10:56:37,015][transformers.modeling_utils][WARNING] - Some weights of Phi3ForSequenceClassification were not initialized from the model checkpoint at microsoft/phi-4 and are newly initialized: ['score.weight'] | |
| You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. | |
| [2025-06-30 10:56:38,598][__main__][INFO] - Initialized new PEFT model for ce loss | |
| [2025-06-30 10:56:38,601][__main__][INFO] - None | |
| [2025-06-30 10:56:38,601][transformers.training_args][INFO] - PyTorch: setting up devices | |
| [2025-06-30 10:56:38,636][__main__][INFO] - Total steps: 620. Number of warmup steps: 62 | |
| [2025-06-30 10:56:38,642][transformers.trainer][INFO] - You have loaded a model on multiple GPUs. `is_model_parallel` attribute will be force-set to `True` to avoid any unexpected behavior such as device placement mismatching. | |
| [2025-06-30 10:56:38,666][transformers.trainer][INFO] - Using auto half precision backend | |
| [2025-06-30 10:56:38,667][transformers.trainer][WARNING] - No label_names provided for model class `PeftModelForSequenceClassification`. Since `PeftModel` hides base models input arguments, if label_names is not given, label_names can't be set automatically within `Trainer`. Note that empty label_names list will be used instead. | |
| [2025-06-30 10:56:38,668][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 10:56:38,682][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 10:56:38,682][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 10:56:38,682][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 10:57:18,125][transformers.trainer][INFO] - The following columns in the Training set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 10:57:18,167][transformers.trainer][INFO] - ***** Running training ***** | |
| [2025-06-30 10:57:18,167][transformers.trainer][INFO] - Num examples = 500 | |
| [2025-06-30 10:57:18,167][transformers.trainer][INFO] - Num Epochs = 20 | |
| [2025-06-30 10:57:18,167][transformers.trainer][INFO] - Instantaneous batch size per device = 1 | |
| [2025-06-30 10:57:18,168][transformers.trainer][INFO] - Total train batch size (w. parallel, distributed & accumulation) = 16 | |
| [2025-06-30 10:57:18,168][transformers.trainer][INFO] - Gradient Accumulation steps = 16 | |
| [2025-06-30 10:57:18,168][transformers.trainer][INFO] - Total optimization steps = 640 | |
| [2025-06-30 10:57:18,170][transformers.trainer][INFO] - Number of trainable parameters = 27,883,520 | |
| [2025-06-30 11:03:17,958][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:03:17,961][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:03:17,961][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:03:17,961][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:03:57,057][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-32 | |
| [2025-06-30 11:03:57,830][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:03:57,831][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:09:58,016][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:09:58,020][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:09:58,020][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:09:58,020][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:10:37,086][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-64 | |
| [2025-06-30 11:10:37,617][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:10:37,617][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:10:38,202][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-32] due to args.save_total_limit | |
| [2025-06-30 11:16:37,828][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:16:37,831][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:16:37,832][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:16:37,832][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:17:16,887][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-96 | |
| [2025-06-30 11:17:17,973][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:17:17,974][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:17:18,630][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-64] due to args.save_total_limit | |
| [2025-06-30 11:23:18,315][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:23:18,319][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:23:18,319][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:23:18,319][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:23:57,367][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-128 | |
| [2025-06-30 11:23:57,835][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:23:57,836][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:23:58,428][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-96] due to args.save_total_limit | |
| [2025-06-30 11:29:58,326][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:29:58,329][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:29:58,330][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:29:58,330][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:30:37,437][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-160 | |
| [2025-06-30 11:30:37,954][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:30:37,954][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:36:38,177][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:36:38,180][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:36:38,180][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:36:38,180][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:37:17,251][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-192 | |
| [2025-06-30 11:37:17,732][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:37:17,732][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:37:18,320][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-160] due to args.save_total_limit | |
| [2025-06-30 11:43:17,766][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:43:17,769][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:43:17,769][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:43:17,769][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:43:56,820][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-224 | |
| [2025-06-30 11:43:57,345][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:43:57,346][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:43:57,936][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-192] due to args.save_total_limit | |
| [2025-06-30 11:49:57,587][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:49:57,590][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:49:57,590][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:49:57,590][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:50:36,688][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-256 | |
| [2025-06-30 11:50:37,177][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:50:37,177][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:50:37,861][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-224] due to args.save_total_limit | |
| [2025-06-30 11:56:38,122][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:56:38,126][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:56:38,126][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:56:38,126][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:57:17,225][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-288 | |
| [2025-06-30 11:57:17,709][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:57:17,710][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:57:18,325][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-256] due to args.save_total_limit | |
| [2025-06-30 11:57:18,349][transformers.trainer][INFO] - | |
| Training completed. Do not forget to share your model on huggingface.co/models =) | |
| [2025-06-30 11:57:18,349][transformers.trainer][INFO] - Loading best model from /workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-128 (score: 0.43522906793048977). | |
| [2025-06-30 11:57:18,580][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-06-30/10-56-18/results/checkpoint-288] due to args.save_total_limit | |
| [2025-06-30 11:57:18,604][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:57:18,610][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:57:18,610][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-06-30 11:57:18,610][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:57:57,713][__main__][INFO] - Training completed successfully. | |
| [2025-06-30 11:57:57,713][__main__][INFO] - Running on Test | |
| [2025-06-30 11:57:57,713][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades. If supporting_text, reference, essay_text, id_prompt, id, essay_year, prompt, grades are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-06-30 11:57:57,717][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-06-30 11:57:57,717][transformers.trainer][INFO] - Num examples = 138 | |
| [2025-06-30 11:57:57,717][transformers.trainer][INFO] - Batch size = 16 | |
| [2025-06-30 11:58:37,928][__main__][INFO] - Test metrics: {'eval_loss': 1.8441004753112793, 'eval_model_preparation_time': 0.0102, 'eval_accuracy': 0.7318840579710145, 'eval_RMSE': 23.59071298478354, 'eval_QWK': 0.5601593625498008, 'eval_HDIV': 0.007246376811594235, 'eval_Macro_F1': 0.25666666666666665, 'eval_Micro_F1': 0.7318840579710145, 'eval_Weighted_F1': 0.6872463768115942, 'eval_TP_0': 0, 'eval_TN_0': 137, 'eval_FP_0': 0, 'eval_FN_0': 1, 'eval_TP_1': 0, 'eval_TN_1': 137, 'eval_FP_1': 0, 'eval_FN_1': 1, 'eval_TP_2': 0, 'eval_TN_2': 129, 'eval_FP_2': 0, 'eval_FN_2': 9, 'eval_TP_3': 64, 'eval_TN_3': 42, 'eval_FP_3': 20, 'eval_FN_3': 12, 'eval_TP_4': 37, 'eval_TN_4': 75, 'eval_FP_4': 17, 'eval_FN_4': 9, 'eval_TP_5': 0, 'eval_TN_5': 133, 'eval_FP_5': 0, 'eval_FN_5': 5, 'eval_runtime': 40.2013, 'eval_samples_per_second': 3.433, 'eval_steps_per_second': 0.224, 'epoch': 9.0} | |
| [2025-06-30 11:58:37,929][transformers.trainer][INFO] - Saving model checkpoint to ./results/best_model | |
| [2025-06-30 11:58:38,392][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-06-30 11:58:38,392][transformers.configuration_utils][INFO] - Model config Phi3Config { | |
| "architectures": [ | |
| "Phi3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100257, | |
| "embd_pdrop": 0.0, | |
| "eos_token_id": 100265, | |
| "hidden_act": "silu", | |
| "hidden_size": 5120, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 17920, | |
| "max_position_embeddings": 16384, | |
| "model_type": "phi3", | |
| "num_attention_heads": 40, | |
| "num_hidden_layers": 40, | |
| "num_key_value_heads": 10, | |
| "original_max_position_embeddings": 16384, | |
| "pad_token_id": 100349, | |
| "partial_rotary_factor": 1.0, | |
| "resid_pdrop": 0.0, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 250000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.53.0", | |
| "use_cache": true, | |
| "vocab_size": 100352 | |
| } | |
| [2025-06-30 11:58:38,812][transformers.tokenization_utils_base][INFO] - chat template saved in ./results/best_model/chat_template.jinja | |
| [2025-06-30 11:58:38,813][transformers.tokenization_utils_base][INFO] - tokenizer config file saved in ./results/best_model/tokenizer_config.json | |
| [2025-06-30 11:58:38,813][transformers.tokenization_utils_base][INFO] - Special tokens file saved in ./results/best_model/special_tokens_map.json | |
| [2025-06-30 11:58:38,891][__main__][INFO] - Model and tokenizer saved to ./results/best_model | |
| [2025-06-30 11:58:38,913][__main__][INFO] - Fine Tuning Finished. | |
| [2025-06-30 11:58:39,421][__main__][INFO] - Total emissions: 0.3926 kg CO2eq | |