| [2025-07-05 02:30:38,581][__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/llama31-8b-balanced/C5 | |
| logging_dir: ./logs/llama31-8b-balanced/C5 | |
| best_model_dir: ./results/llama31-8b-balanced/C5/best_model | |
| lora_r: 8 | |
| lora_dropout: 0.05 | |
| lora_alpha: 16 | |
| lora_target_modules: all-linear | |
| tokenizer: | |
| name: microsoft/phi-4 | |
| dataset: | |
| grade_index: 4 | |
| 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 | |
| [2025-07-05 02:30:42,515][__main__][INFO] - GPU 0: NVIDIA H200 | TDP ≈ 700 W | |
| [2025-07-05 02:30:42,515][__main__][INFO] - Starting the Fine Tuning training process. | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file vocab.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/vocab.json | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file merges.txt from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/merges.txt | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file tokenizer.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer.json | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file added_tokens.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/added_tokens.json | |
| [2025-07-05 02:30:46,288][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 | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file tokenizer_config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer_config.json | |
| [2025-07-05 02:30:46,288][transformers.tokenization_utils_base][INFO] - loading file chat_template.jinja from cache at None | |
| [2025-07-05 02:30:46,425][__main__][INFO] - Tokenizer function parameters- Padding:longest; Truncation: False; Use Full Context: True | |
| [2025-07-05 02:30:48,049][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 02:30:48,050][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 | |
| } | |
| [2025-07-05 02:30:48,177][transformers.modeling_utils][INFO] - loading weights file model.safetensors from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/model.safetensors.index.json | |
| [2025-07-05 02:30:48,178][transformers.modeling_utils][INFO] - Will use torch_dtype=torch.bfloat16 as defined in model's config object | |
| [2025-07-05 02:30:48,178][transformers.modeling_utils][INFO] - Instantiating Phi3ForSequenceClassification model under default dtype torch.bfloat16. | |
| [2025-07-05 02:31:06,485][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). | |
| [2025-07-05 02:31:06,485][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. | |
| [2025-07-05 02:31:07,659][__main__][INFO] - Initialized new PEFT model for ce loss | |
| [2025-07-05 02:31:07,661][__main__][INFO] - None | |
| [2025-07-05 02:31:07,662][transformers.training_args][INFO] - PyTorch: setting up devices | |
| [2025-07-05 02:31:07,698][__main__][INFO] - Total steps: 620. Number of warmup steps: 62 | |
| [2025-07-05 02:31:07,705][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. | |
| [2025-07-05 02:31:07,738][transformers.trainer][INFO] - Using auto half precision backend | |
| [2025-07-05 02:31:07,738][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. | |
| [2025-07-05 02:31:07,740][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 02:31:07,752][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 02:31:07,752][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 02:31:07,752][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 02:31:48,935][transformers.trainer][INFO] - The following columns in the Training set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - ***** Running training ***** | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Num examples = 500 | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Num Epochs = 20 | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Instantaneous batch size per device = 8 | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Total train batch size (w. parallel, distributed & accumulation) = 16 | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Gradient Accumulation steps = 2 | |
| [2025-07-05 02:31:48,975][transformers.trainer][INFO] - Total optimization steps = 640 | |
| [2025-07-05 02:31:48,977][transformers.trainer][INFO] - Number of trainable parameters = 27,883,520 | |
| [2025-07-05 02:31:49,062][transformers.models.phi3.modeling_phi3][WARNING] - `use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`. | |
| [2025-07-05 02:40:22,627][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 02:40:22,630][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 02:40:22,630][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 02:40:22,631][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 02:41:03,611][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-32 | |
| [2025-07-05 02:41:03,981][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 02:41:03,981][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 | |
| } | |
| [2025-07-05 02:49:37,919][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 02:49:37,922][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 02:49:37,923][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 02:49:37,923][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 02:50:18,865][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-64 | |
| [2025-07-05 02:50:19,525][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 02:50:19,525][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 | |
| } | |
| [2025-07-05 02:50:20,115][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-32] due to args.save_total_limit | |
| [2025-07-05 02:58:53,656][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 02:58:53,659][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 02:58:53,660][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 02:58:53,660][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 02:59:34,676][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-96 | |
| [2025-07-05 02:59:35,168][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 02:59:35,168][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 | |
| } | |
| [2025-07-05 03:08:09,268][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:08:09,272][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:08:09,272][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:08:09,272][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:08:50,301][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-128 | |
| [2025-07-05 03:08:50,629][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:08:50,629][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 | |
| } | |
| [2025-07-05 03:08:51,180][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-64] due to args.save_total_limit | |
| [2025-07-05 03:08:51,200][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-96] due to args.save_total_limit | |
| [2025-07-05 03:17:24,115][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:17:24,119][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:17:24,119][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:17:24,119][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:18:05,082][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-160 | |
| [2025-07-05 03:18:05,414][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:18:05,415][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 | |
| } | |
| [2025-07-05 03:18:06,013][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-128] due to args.save_total_limit | |
| [2025-07-05 03:26:39,741][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:26:39,744][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:26:39,745][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:26:39,745][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:27:20,716][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-192 | |
| [2025-07-05 03:27:21,233][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:27:21,234][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 | |
| } | |
| [2025-07-05 03:27:21,784][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-160] due to args.save_total_limit | |
| [2025-07-05 03:35:55,368][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:35:55,371][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:35:55,371][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:35:55,371][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:36:36,371][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-224 | |
| [2025-07-05 03:36:36,709][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:36:36,710][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 | |
| } | |
| [2025-07-05 03:36:37,312][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-192] due to args.save_total_limit | |
| [2025-07-05 03:45:10,666][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:45:10,670][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:45:10,670][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:45:10,670][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:45:51,676][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-256 | |
| [2025-07-05 03:45:52,185][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:45:52,185][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 | |
| } | |
| [2025-07-05 03:54:26,388][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 03:54:26,391][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 03:54:26,391][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 03:54:26,391][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 03:55:07,400][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-288 | |
| [2025-07-05 03:55:07,894][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 03:55:07,894][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 | |
| } | |
| [2025-07-05 03:55:08,405][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-224] due to args.save_total_limit | |
| [2025-07-05 03:55:08,426][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-256] due to args.save_total_limit | |
| [2025-07-05 04:03:42,219][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:03:42,223][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:03:42,223][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:03:42,223][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:04:23,234][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-320 | |
| [2025-07-05 04:04:23,757][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:04:23,757][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 | |
| } | |
| [2025-07-05 04:12:58,078][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:12:58,081][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:12:58,081][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:12:58,081][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:13:39,078][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-352 | |
| [2025-07-05 04:13:39,401][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:13:39,402][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 | |
| } | |
| [2025-07-05 04:13:39,876][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-320] due to args.save_total_limit | |
| [2025-07-05 04:22:13,770][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:22:13,774][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:22:13,774][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:22:13,774][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:22:54,740][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-384 | |
| [2025-07-05 04:22:55,077][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:22:55,077][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 | |
| } | |
| [2025-07-05 04:22:55,578][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-352] due to args.save_total_limit | |
| [2025-07-05 04:31:29,090][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:31:29,094][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:31:29,094][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:31:29,094][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:32:10,085][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-416 | |
| [2025-07-05 04:32:10,421][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:32:10,421][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 | |
| } | |
| [2025-07-05 04:32:10,976][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-384] due to args.save_total_limit | |
| [2025-07-05 04:40:44,811][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:40:44,814][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:40:44,815][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:40:44,815][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:41:25,805][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-448 | |
| [2025-07-05 04:41:26,374][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:41:26,374][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 | |
| } | |
| [2025-07-05 04:41:26,882][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-416] due to args.save_total_limit | |
| [2025-07-05 04:41:26,902][transformers.trainer][INFO] - | |
| Training completed. Do not forget to share your model on huggingface.co/models =) | |
| [2025-07-05 04:41:26,902][transformers.trainer][INFO] - Loading best model from /workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-288 (score: 0.5763501435458533). | |
| [2025-07-05 04:41:26,968][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-05/02-30-38/results/llama31-8b-balanced/C5/checkpoint-448] due to args.save_total_limit | |
| [2025-07-05 04:41:26,992][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:41:26,997][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:41:26,997][transformers.trainer][INFO] - Num examples = 132 | |
| [2025-07-05 04:41:26,998][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:42:07,984][__main__][INFO] - Training completed successfully. | |
| [2025-07-05 04:42:07,984][__main__][INFO] - Running on Test | |
| [2025-07-05 04:42:07,984][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt. If grades, essay_text, essay_year, reference, id, supporting_text, prompt, id_prompt are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message. | |
| [2025-07-05 04:42:07,987][transformers.trainer][INFO] - | |
| ***** Running Evaluation ***** | |
| [2025-07-05 04:42:07,988][transformers.trainer][INFO] - Num examples = 138 | |
| [2025-07-05 04:42:07,988][transformers.trainer][INFO] - Batch size = 4 | |
| [2025-07-05 04:42:51,285][__main__][INFO] - Test metrics: {'eval_loss': 1.5521295070648193, 'eval_model_preparation_time': 0.0089, 'eval_accuracy': 0.30434782608695654, 'eval_RMSE': 57.4834569775249, 'eval_QWK': 0.5246440571442383, 'eval_HDIV': 0.050724637681159424, 'eval_Macro_F1': 0.2966523012883595, 'eval_Micro_F1': 0.30434782608695654, 'eval_Weighted_F1': 0.3060776025506538, 'eval_TP_0': 15, 'eval_TN_0': 109, 'eval_FP_0': 7, 'eval_FN_0': 7, 'eval_TP_1': 3, 'eval_TN_1': 97, 'eval_FP_1': 9, 'eval_FN_1': 29, 'eval_TP_2': 8, 'eval_TN_2': 96, 'eval_FP_2': 18, 'eval_FN_2': 16, 'eval_TP_3': 8, 'eval_TN_3': 70, 'eval_FP_3': 43, 'eval_FN_3': 17, 'eval_TP_4': 7, 'eval_TN_4': 98, 'eval_FP_4': 8, 'eval_FN_4': 25, 'eval_TP_5': 1, 'eval_TN_5': 124, 'eval_FP_5': 11, 'eval_FN_5': 2, 'eval_runtime': 43.2885, 'eval_samples_per_second': 3.188, 'eval_steps_per_second': 0.809, 'epoch': 14.0} | |
| [2025-07-05 04:42:51,286][transformers.trainer][INFO] - Saving model checkpoint to ./results/llama31-8b-balanced/C5/best_model | |
| [2025-07-05 04:42:51,592][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json | |
| [2025-07-05 04:42:51,592][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 | |
| } | |
| [2025-07-05 04:42:51,883][transformers.tokenization_utils_base][INFO] - chat template saved in ./results/llama31-8b-balanced/C5/best_model/chat_template.jinja | |
| [2025-07-05 04:42:51,884][transformers.tokenization_utils_base][INFO] - tokenizer config file saved in ./results/llama31-8b-balanced/C5/best_model/tokenizer_config.json | |
| [2025-07-05 04:42:51,884][transformers.tokenization_utils_base][INFO] - Special tokens file saved in ./results/llama31-8b-balanced/C5/best_model/special_tokens_map.json | |
| [2025-07-05 04:42:51,952][__main__][INFO] - Model and tokenizer saved to ./results/llama31-8b-balanced/C5/best_model | |
| [2025-07-05 04:42:51,977][__main__][INFO] - Fine Tuning Finished. | |
| [2025-07-05 04:42:52,490][__main__][INFO] - Total emissions: 0.1012 kg CO2eq | |