| [2025-07-06 08:03:18,144][__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/phi4-balanced/C3 |
| logging_dir: ./logs/phi4-balanced/C3 |
| best_model_dir: ./results/phi4-balanced/C3/best_model |
| lora_r: 16 |
| lora_dropout: 0.1 |
| lora_alpha: 32 |
| lora_target_modules: all-linear |
| tokenizer: |
| name: microsoft/phi-4 |
| dataset: |
| grade_index: 2 |
| use_full_context: true |
| training_params: |
| weight_decay: 0.01 |
| warmup_ratio: 0.1 |
| learning_rate: 5.0e-05 |
| train_batch_size: 8 |
| eval_batch_size: 4 |
| gradient_accumulation_steps: 2 |
| gradient_checkpointing: true |
|
|
| [2025-07-06 08:03:22,029][__main__][INFO] - GPU 0: NVIDIA H200 | TDP ≈ 700 W |
| [2025-07-06 08:03:22,029][__main__][INFO] - Starting the Fine Tuning training process. |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file vocab.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/vocab.json |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file merges.txt from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/merges.txt |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file tokenizer.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer.json |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file added_tokens.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/added_tokens.json |
| [2025-07-06 08:03:26,067][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 |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file tokenizer_config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer_config.json |
| [2025-07-06 08:03:26,067][transformers.tokenization_utils_base][INFO] - loading file chat_template.jinja from cache at None |
| [2025-07-06 08:03:26,228][__main__][INFO] - Tokenizer function parameters- Padding:longest; Truncation: False; Use Full Context: True |
| [2025-07-06 08:03:26,714][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:03:26,714][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 |
| } |
|
|
| [2025-07-06 08:03:26,846][transformers.modeling_utils][INFO] - loading weights file model.safetensors from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/model.safetensors.index.json |
| [2025-07-06 08:03:26,847][transformers.modeling_utils][INFO] - Will use torch_dtype=torch.bfloat16 as defined in model's config object |
| [2025-07-06 08:03:26,847][transformers.modeling_utils][INFO] - Instantiating Phi3ForSequenceClassification model under default dtype torch.bfloat16. |
| [2025-07-06 08:03:44,376][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). |
| [2025-07-06 08:03:44,376][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. |
| [2025-07-06 08:03:46,965][__main__][INFO] - Initialized new PEFT model for ce loss |
| [2025-07-06 08:03:46,967][__main__][INFO] - None |
| [2025-07-06 08:03:46,969][transformers.training_args][INFO] - PyTorch: setting up devices |
| [2025-07-06 08:03:47,006][__main__][INFO] - Total steps: 620. Number of warmup steps: 62 |
| [2025-07-06 08:03:47,014][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. |
| [2025-07-06 08:03:47,048][transformers.trainer][INFO] - Using auto half precision backend |
| [2025-07-06 08:03:47,048][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. |
| [2025-07-06 08:03:47,049][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:03:47,062][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:03:47,062][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:03:47,062][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:04:26,999][transformers.trainer][INFO] - The following columns in the Training set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - ***** Running training ***** |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Num examples = 500 |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Num Epochs = 20 |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Instantaneous batch size per device = 8 |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Total train batch size (w. parallel, distributed & accumulation) = 16 |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Gradient Accumulation steps = 2 |
| [2025-07-06 08:04:27,040][transformers.trainer][INFO] - Total optimization steps = 640 |
| [2025-07-06 08:04:27,042][transformers.trainer][INFO] - Number of trainable parameters = 55,736,320 |
| [2025-07-06 08:04:27,124][transformers.models.phi3.modeling_phi3][WARNING] - `use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`. |
| [2025-07-06 08:12:36,207][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:12:36,210][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:12:36,210][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:12:36,210][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:13:15,778][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-32 |
| [2025-07-06 08:13:16,107][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:13:16,108][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 |
| } |
| |
| [2025-07-06 08:21:26,562][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:21:26,566][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:21:26,566][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:21:26,566][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:22:06,101][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-64 |
| [2025-07-06 08:22:06,594][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:22:06,595][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 |
| } |
|
|
| [2025-07-06 08:22:07,661][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-32] due to args.save_total_limit |
| [2025-07-06 08:30:17,223][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:30:17,226][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:30:17,227][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:30:17,227][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:30:56,773][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-96 |
| [2025-07-06 08:30:57,112][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:30:57,113][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 |
| } |
| |
| [2025-07-06 08:30:58,169][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-64] due to args.save_total_limit |
| [2025-07-06 08:39:08,698][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:39:08,702][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:39:08,702][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:39:08,702][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:39:48,307][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-128 |
| [2025-07-06 08:39:48,643][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:39:48,643][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 |
| } |
|
|
| [2025-07-06 08:47:59,549][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:47:59,553][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:47:59,553][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:47:59,553][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:48:39,147][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-160 |
| [2025-07-06 08:48:39,485][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:48:39,485][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 |
| } |
| |
| [2025-07-06 08:48:40,570][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-96] due to args.save_total_limit |
| [2025-07-06 08:48:40,603][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-128] due to args.save_total_limit |
| [2025-07-06 08:56:50,236][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 08:56:50,240][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 08:56:50,240][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 08:56:50,240][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 08:57:29,800][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-192 |
| [2025-07-06 08:57:30,143][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 08:57:30,144][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 |
| } |
|
|
| [2025-07-06 08:57:31,272][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-160] due to args.save_total_limit |
| [2025-07-06 09:05:41,198][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:05:41,201][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:05:41,201][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:05:41,201][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:06:20,820][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-224 |
| [2025-07-06 09:06:21,145][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:06:21,146][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 |
| } |
| |
| [2025-07-06 09:14:32,052][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:14:32,055][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:14:32,055][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:14:32,055][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:15:11,644][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-256 |
| [2025-07-06 09:15:12,038][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:15:12,039][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 |
| } |
|
|
| [2025-07-06 09:15:13,092][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-224] due to args.save_total_limit |
| [2025-07-06 09:23:22,593][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:23:22,596][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:23:22,597][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:23:22,597][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:24:02,168][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-288 |
| [2025-07-06 09:24:02,481][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:24:02,481][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 |
| } |
| |
| [2025-07-06 09:24:03,585][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-256] due to args.save_total_limit |
| [2025-07-06 09:32:13,182][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:32:13,185][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:32:13,185][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:32:13,185][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:32:52,779][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-320 |
| [2025-07-06 09:32:53,275][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:32:53,276][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 |
| } |
|
|
| [2025-07-06 09:32:54,450][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-192] due to args.save_total_limit |
| [2025-07-06 09:32:54,492][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-288] due to args.save_total_limit |
| [2025-07-06 09:41:04,041][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:41:04,044][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:41:04,045][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:41:04,045][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:41:43,585][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-352 |
| [2025-07-06 09:41:44,083][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:41:44,083][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 |
| } |
| |
| [2025-07-06 09:49:54,606][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:49:54,609][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:49:54,610][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:49:54,610][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:50:34,224][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-384 |
| [2025-07-06 09:50:34,552][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:50:34,553][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 |
| } |
|
|
| [2025-07-06 09:50:35,697][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-352] due to args.save_total_limit |
| [2025-07-06 09:58:45,644][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 09:58:45,648][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 09:58:45,648][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 09:58:45,648][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 09:59:25,244][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-416 |
| [2025-07-06 09:59:25,570][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 09:59:25,570][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 |
| } |
| |
| [2025-07-06 09:59:26,588][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-384] due to args.save_total_limit |
| [2025-07-06 10:07:36,313][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 10:07:36,317][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 10:07:36,317][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 10:07:36,317][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 10:08:15,871][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-448 |
| [2025-07-06 10:08:16,232][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 10:08:16,233][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 |
| } |
|
|
| [2025-07-06 10:08:17,391][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-416] due to args.save_total_limit |
| [2025-07-06 10:16:27,127][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 10:16:27,131][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 10:16:27,131][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 10:16:27,131][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 10:17:06,707][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-480 |
| [2025-07-06 10:17:07,031][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 10:17:07,032][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 |
| } |
| |
| [2025-07-06 10:17:08,169][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-448] due to args.save_total_limit |
| [2025-07-06 10:17:08,210][transformers.trainer][INFO] - |
| |
| Training completed. Do not forget to share your model on huggingface.co/models =) |
| |
| |
| [2025-07-06 10:17:08,210][transformers.trainer][INFO] - Loading best model from /workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-320 (score: 0.6056215783148617). |
| [2025-07-06 10:17:08,478][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-06/08-03-18/results/phi4-balanced/C3/checkpoint-480] due to args.save_total_limit |
| [2025-07-06 10:17:08,527][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 10:17:08,536][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 10:17:08,536][transformers.trainer][INFO] - Num examples = 132 |
| [2025-07-06 10:17:08,536][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 10:17:48,131][__main__][INFO] - Training completed successfully. |
| [2025-07-06 10:17:48,131][__main__][INFO] - Running on Test |
| [2025-07-06 10:17:48,131][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year. If essay_text, supporting_text, grades, prompt, id, reference, id_prompt, essay_year are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. |
| [2025-07-06 10:17:48,134][transformers.trainer][INFO] - |
| ***** Running Evaluation ***** |
| [2025-07-06 10:17:48,134][transformers.trainer][INFO] - Num examples = 138 |
| [2025-07-06 10:17:48,134][transformers.trainer][INFO] - Batch size = 4 |
| [2025-07-06 10:18:29,986][__main__][INFO] - Test metrics: {'eval_loss': 1.3064757585525513, 'eval_model_preparation_time': 0.0091, 'eval_accuracy': 0.4492753623188406, 'eval_RMSE': 44.78612621615399, 'eval_QWK': 0.6485913627130618, 'eval_HDIV': 0.02898550724637683, 'eval_Macro_F1': 0.3651102312304498, 'eval_Micro_F1': 0.4492753623188406, 'eval_Weighted_F1': 0.45017539006610047, 'eval_TP_0': 0, 'eval_TN_0': 137, 'eval_FP_0': 0, 'eval_FN_0': 1, 'eval_TP_1': 19, 'eval_TN_1': 96, 'eval_FP_1': 13, 'eval_FN_1': 10, 'eval_TP_2': 9, 'eval_TN_2': 105, 'eval_FP_2': 15, 'eval_FN_2': 9, 'eval_TP_3': 18, 'eval_TN_3': 75, 'eval_FP_3': 18, 'eval_FN_3': 27, 'eval_TP_4': 10, 'eval_TN_4': 94, 'eval_FP_4': 6, 'eval_FN_4': 28, 'eval_TP_5': 6, 'eval_TN_5': 107, 'eval_FP_5': 24, 'eval_FN_5': 1, 'eval_runtime': 41.8421, 'eval_samples_per_second': 3.298, 'eval_steps_per_second': 0.836, 'epoch': 15.0} |
| [2025-07-06 10:18:29,986][transformers.trainer][INFO] - Saving model checkpoint to ./results/phi4-balanced/C3/best_model |
| [2025-07-06 10:18:30,293][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json |
| [2025-07-06 10:18:30,293][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 |
| } |
| |
| [2025-07-06 10:18:31,033][transformers.tokenization_utils_base][INFO] - chat template saved in ./results/phi4-balanced/C3/best_model/chat_template.jinja |
| [2025-07-06 10:18:31,034][transformers.tokenization_utils_base][INFO] - tokenizer config file saved in ./results/phi4-balanced/C3/best_model/tokenizer_config.json |
| [2025-07-06 10:18:31,034][transformers.tokenization_utils_base][INFO] - Special tokens file saved in ./results/phi4-balanced/C3/best_model/special_tokens_map.json |
| [2025-07-06 10:18:31,111][__main__][INFO] - Model and tokenizer saved to ./results/phi4-balanced/C3/best_model |
| [2025-07-06 10:18:31,132][__main__][INFO] - Fine Tuning Finished. |
| [2025-07-06 10:18:31,641][__main__][INFO] - Total emissions: 0.1035 kg CO2eq |
| |
