abarbosa's picture
Pushing fine-tuned model to Hugging Face Hub
c602627 verified
[2025-07-04 14:50:14,678][__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/phi35-balanced/C4
logging_dir: ./logs/phi35-balanced/C4
best_model_dir: ./results/phi35-balanced/C4/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: 3
use_full_context: false
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
[2025-07-04 14:50:18,576][__main__][INFO] - GPU 0: NVIDIA H200 | TDP ≈ 700 W
[2025-07-04 14:50:18,576][__main__][INFO] - Starting the Fine Tuning training process.
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file vocab.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/vocab.json
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file merges.txt from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/merges.txt
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file tokenizer.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer.json
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file added_tokens.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/added_tokens.json
[2025-07-04 14:50:21,972][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
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file tokenizer_config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/tokenizer_config.json
[2025-07-04 14:50:21,972][transformers.tokenization_utils_base][INFO] - loading file chat_template.jinja from cache at None
[2025-07-04 14:50:22,110][__main__][INFO] - Tokenizer function parameters- Padding:longest; Truncation: False; Use Full Context: False
[2025-07-04 14:50:23,226][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 14:50:23,227][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
}
[2025-07-04 14:50:23,354][transformers.modeling_utils][INFO] - loading weights file model.safetensors from cache at /tmp/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/model.safetensors.index.json
[2025-07-04 14:50:23,354][transformers.modeling_utils][INFO] - Will use torch_dtype=torch.bfloat16 as defined in model's config object
[2025-07-04 14:50:23,354][transformers.modeling_utils][INFO] - Instantiating Phi3ForSequenceClassification model under default dtype torch.bfloat16.
[2025-07-04 14:50:41,325][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).
[2025-07-04 14:50:41,325][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.
[2025-07-04 14:50:42,871][__main__][INFO] - Initialized new PEFT model for ce loss
[2025-07-04 14:50:42,873][__main__][INFO] - None
[2025-07-04 14:50:42,874][transformers.training_args][INFO] - PyTorch: setting up devices
[2025-07-04 14:50:42,910][__main__][INFO] - Total steps: 620. Number of warmup steps: 62
[2025-07-04 14:50:42,917][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.
[2025-07-04 14:50:42,951][transformers.trainer][INFO] - Using auto half precision backend
[2025-07-04 14:50:42,951][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.
[2025-07-04 14:50:42,953][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 14:50:42,966][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 14:50:42,966][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 14:50:42,966][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 14:51:07,361][transformers.trainer][INFO] - The following columns in the Training set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 14:51:07,401][transformers.trainer][INFO] - ***** Running training *****
[2025-07-04 14:51:07,401][transformers.trainer][INFO] - Num examples = 500
[2025-07-04 14:51:07,402][transformers.trainer][INFO] - Num Epochs = 20
[2025-07-04 14:51:07,402][transformers.trainer][INFO] - Instantaneous batch size per device = 8
[2025-07-04 14:51:07,402][transformers.trainer][INFO] - Total train batch size (w. parallel, distributed & accumulation) = 16
[2025-07-04 14:51:07,402][transformers.trainer][INFO] - Gradient Accumulation steps = 2
[2025-07-04 14:51:07,402][transformers.trainer][INFO] - Total optimization steps = 640
[2025-07-04 14:51:07,404][transformers.trainer][INFO] - Number of trainable parameters = 27,883,520
[2025-07-04 14:51:07,465][transformers.models.phi3.modeling_phi3][WARNING] - `use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
[2025-07-04 14:56:46,999][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 14:56:47,003][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 14:56:47,003][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 14:56:47,003][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 14:57:11,039][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-32
[2025-07-04 14:57:11,389][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 14:57:11,389][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
}
[2025-07-04 15:02:51,424][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:02:51,427][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:02:51,427][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:02:51,428][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:03:15,458][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-64
[2025-07-04 15:03:15,813][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:03:15,813][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
}
[2025-07-04 15:08:55,694][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:08:55,697][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:08:55,697][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:08:55,697][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:09:19,709][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-96
[2025-07-04 15:09:20,053][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:09:20,054][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
}
[2025-07-04 15:09:20,584][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-64] due to args.save_total_limit
[2025-07-04 15:14:59,794][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:14:59,798][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:14:59,798][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:14:59,798][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:15:23,661][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-128
[2025-07-04 15:15:24,022][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:15:24,023][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
}
[2025-07-04 15:15:24,584][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-32] due to args.save_total_limit
[2025-07-04 15:15:24,604][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-96] due to args.save_total_limit
[2025-07-04 15:21:02,887][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:21:02,891][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:21:02,891][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:21:02,891][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:21:26,892][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-160
[2025-07-04 15:21:27,262][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:21:27,263][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
}
[2025-07-04 15:27:06,614][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:27:06,617][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:27:06,617][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:27:06,617][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:27:30,612][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-192
[2025-07-04 15:27:30,953][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:27:30,953][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
}
[2025-07-04 15:27:31,469][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-160] due to args.save_total_limit
[2025-07-04 15:33:10,344][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:33:10,347][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:33:10,347][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:33:10,347][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:33:34,337][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-224
[2025-07-04 15:33:35,000][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:33:35,001][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
}
[2025-07-04 15:33:35,566][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-192] due to args.save_total_limit
[2025-07-04 15:39:14,434][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:39:14,437][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:39:14,437][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:39:14,438][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:39:38,434][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-256
[2025-07-04 15:39:38,750][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:39:38,750][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
}
[2025-07-04 15:39:39,314][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-224] due to args.save_total_limit
[2025-07-04 15:45:18,242][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:45:18,245][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:45:18,245][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:45:18,245][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:45:42,251][transformers.trainer][INFO] - Saving model checkpoint to /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-288
[2025-07-04 15:45:42,607][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:45:42,607][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
}
[2025-07-04 15:45:43,100][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-256] due to args.save_total_limit
[2025-07-04 15:45:43,114][transformers.trainer][INFO] -
Training completed. Do not forget to share your model on huggingface.co/models =)
[2025-07-04 15:45:43,114][transformers.trainer][INFO] - Loading best model from /workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-128 (score: 0.5317662007623889).
[2025-07-04 15:45:43,253][transformers.trainer][INFO] - Deleting older checkpoint [/workspace/jbcs2025/outputs/2025-07-04/14-50-14/results/phi35-balanced/C4/checkpoint-288] due to args.save_total_limit
[2025-07-04 15:45:43,271][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:45:43,277][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:45:43,277][transformers.trainer][INFO] - Num examples = 132
[2025-07-04 15:45:43,277][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:46:07,262][__main__][INFO] - Training completed successfully.
[2025-07-04 15:46:07,262][__main__][INFO] - Running on Test
[2025-07-04 15:46:07,263][transformers.trainer][INFO] - The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForSequenceClassification.forward` and have been ignored: id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text. If id_prompt, grades, id, supporting_text, prompt, reference, essay_year, essay_text are not expected by `PeftModelForSequenceClassification.forward`, you can safely ignore this message.
[2025-07-04 15:46:07,265][transformers.trainer][INFO] -
***** Running Evaluation *****
[2025-07-04 15:46:07,266][transformers.trainer][INFO] - Num examples = 138
[2025-07-04 15:46:07,266][transformers.trainer][INFO] - Batch size = 4
[2025-07-04 15:46:32,813][__main__][INFO] - Test metrics: {'eval_loss': 1.1784944534301758, 'eval_model_preparation_time': 0.0091, 'eval_accuracy': 0.6014492753623188, 'eval_RMSE': 27.662562992324986, 'eval_QWK': 0.4293233082706768, 'eval_HDIV': 0.007246376811594235, 'eval_Macro_F1': 0.1934934934934935, 'eval_Micro_F1': 0.6014492753623188, 'eval_Weighted_F1': 0.5512381947164555, 'eval_TP_0': 0, 'eval_TN_0': 137, 'eval_FP_0': 0, 'eval_FN_0': 1, 'eval_TP_1': 0, 'eval_TN_1': 137, 'eval_FP_1': 0, 'eval_FN_1': 1, 'eval_TP_2': 0, 'eval_TN_2': 129, 'eval_FP_2': 0, 'eval_FN_2': 9, 'eval_TP_3': 68, 'eval_TN_3': 26, 'eval_FP_3': 36, 'eval_FN_3': 8, 'eval_TP_4': 15, 'eval_TN_4': 79, 'eval_FP_4': 13, 'eval_FN_4': 31, 'eval_TP_5': 0, 'eval_TN_5': 127, 'eval_FP_5': 6, 'eval_FN_5': 5, 'eval_runtime': 25.5375, 'eval_samples_per_second': 5.404, 'eval_steps_per_second': 1.371, 'epoch': 9.0}
[2025-07-04 15:46:32,813][transformers.trainer][INFO] - Saving model checkpoint to ./results/phi35-balanced/C4/best_model
[2025-07-04 15:46:33,133][transformers.configuration_utils][INFO] - loading configuration file config.json from cache at /workspace/.hf_home/hub/models--microsoft--phi-4/snapshots/187ef0342fff0eb3333be9f00389385e95ef0b61/config.json
[2025-07-04 15:46:33,134][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
}
[2025-07-04 15:46:33,480][transformers.tokenization_utils_base][INFO] - chat template saved in ./results/phi35-balanced/C4/best_model/chat_template.jinja
[2025-07-04 15:46:33,481][transformers.tokenization_utils_base][INFO] - tokenizer config file saved in ./results/phi35-balanced/C4/best_model/tokenizer_config.json
[2025-07-04 15:46:33,481][transformers.tokenization_utils_base][INFO] - Special tokens file saved in ./results/phi35-balanced/C4/best_model/special_tokens_map.json
[2025-07-04 15:46:33,553][__main__][INFO] - Model and tokenizer saved to ./results/phi35-balanced/C4/best_model
[2025-07-04 15:46:33,575][__main__][INFO] - Fine Tuning Finished.
[2025-07-04 15:46:34,084][__main__][INFO] - Total emissions: 0.0432 kg CO2eq