[INFO|2025-02-22 11:20:55] configuration_utils.py:696 >> loading configuration file config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\config.json

[INFO|2025-02-22 11:20:55] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 896,
  "initializer_range": 0.02,
  "intermediate_size": 4864,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 14,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.3",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}


[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file vocab.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\vocab.json

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file merges.txt from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\merges.txt

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file tokenizer.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\tokenizer.json

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file added_tokens.json from cache at None

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file special_tokens_map.json from cache at None

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file tokenizer_config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\tokenizer_config.json

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2034 >> loading file chat_template.jinja from cache at None

[INFO|2025-02-22 11:20:55] tokenization_utils_base.py:2304 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.

[INFO|2025-02-22 11:20:56] configuration_utils.py:696 >> loading configuration file config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\config.json

[INFO|2025-02-22 11:20:56] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 896,
  "initializer_range": 0.02,
  "intermediate_size": 4864,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 14,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.3",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}


[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file vocab.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\vocab.json

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file merges.txt from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\merges.txt

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file tokenizer.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\tokenizer.json

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file added_tokens.json from cache at None

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file special_tokens_map.json from cache at None

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file tokenizer_config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\tokenizer_config.json

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2034 >> loading file chat_template.jinja from cache at None

[INFO|2025-02-22 11:20:56] tokenization_utils_base.py:2304 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.

[INFO|2025-02-22 11:20:56] logging.py:157 >> Add <|im_end|> to stop words.

[INFO|2025-02-22 11:20:56] logging.py:157 >> Loading dataset prithivMLmods/Deepthink-Reasoning-Ins...

[INFO|2025-02-22 11:21:19] logging.py:157 >> Loading dataset open-thoughts/OpenThoughts-114k...

[INFO|2025-02-22 11:27:18] configuration_utils.py:696 >> loading configuration file config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\config.json

[INFO|2025-02-22 11:27:18] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 896,
  "initializer_range": 0.02,
  "intermediate_size": 4864,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 14,
  "num_hidden_layers": 24,
  "num_key_value_heads": 2,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.3",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151936
}


[INFO|2025-02-22 11:27:18] modeling_utils.py:3904 >> loading weights file model.safetensors from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\model.safetensors

[INFO|2025-02-22 11:27:18] modeling_utils.py:1582 >> Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.

[INFO|2025-02-22 11:27:18] configuration_utils.py:1140 >> Generate config GenerationConfig {
  "bos_token_id": 151643,
  "eos_token_id": 151645
}


[INFO|2025-02-22 11:27:21] modeling_utils.py:4888 >> All model checkpoint weights were used when initializing Qwen2ForCausalLM.


[INFO|2025-02-22 11:27:21] modeling_utils.py:4896 >> All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at Qwen/Qwen2.5-0.5B-Instruct.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.

[INFO|2025-02-22 11:27:22] configuration_utils.py:1095 >> loading configuration file generation_config.json from cache at C:\Users\Asad_\.cache\huggingface\hub\models--Qwen--Qwen2.5-0.5B-Instruct\snapshots\7ae557604adf67be50417f59c2c2f167def9a775\generation_config.json

[INFO|2025-02-22 11:27:22] configuration_utils.py:1140 >> Generate config GenerationConfig {
  "bos_token_id": 151643,
  "do_sample": true,
  "eos_token_id": [
    151645,
    151643
  ],
  "pad_token_id": 151643,
  "repetition_penalty": 1.1,
  "temperature": 0.7,
  "top_k": 20,
  "top_p": 0.8
}


[INFO|2025-02-22 11:27:22] logging.py:157 >> Gradient checkpointing enabled.

[INFO|2025-02-22 11:27:22] logging.py:157 >> Using torch SDPA for faster training and inference.

[INFO|2025-02-22 11:27:22] logging.py:157 >> Upcasting trainable params to float32.

[INFO|2025-02-22 11:27:22] logging.py:157 >> Fine-tuning method: Full

[INFO|2025-02-22 11:27:22] logging.py:157 >> trainable params: 494,032,768 || all params: 494,032,768 || trainable%: 100.0000

[INFO|2025-02-22 11:27:22] trainer.py:741 >> Using auto half precision backend

[INFO|2025-02-22 11:27:22] trainer.py:2775 >> Loading model from saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\checkpoint-13000.

[WARNING|2025-02-22 11:27:23] trainer.py:3018 >> There were missing keys in the checkpoint model loaded: ['lm_head.weight'].

[INFO|2025-02-22 11:27:23] trainer.py:2369 >> ***** Running training *****

[INFO|2025-02-22 11:27:23] trainer.py:2370 >>   Num examples = 102,787

[INFO|2025-02-22 11:27:23] trainer.py:2371 >>   Num Epochs = 1

[INFO|2025-02-22 11:27:23] trainer.py:2372 >>   Instantaneous batch size per device = 1

[INFO|2025-02-22 11:27:23] trainer.py:2375 >>   Total train batch size (w. parallel, distributed & accumulation) = 8

[INFO|2025-02-22 11:27:23] trainer.py:2376 >>   Gradient Accumulation steps = 8

[INFO|2025-02-22 11:27:23] trainer.py:2377 >>   Total optimization steps = 12,848

[INFO|2025-02-22 11:27:23] trainer.py:2378 >>   Number of trainable parameters = 494,032,768

[INFO|2025-02-22 11:27:23] trainer.py:2400 >>   Continuing training from checkpoint, will skip to saved global_step

[INFO|2025-02-22 11:27:23] trainer.py:2401 >>   Continuing training from epoch 1

[INFO|2025-02-22 11:27:23] trainer.py:2402 >>   Continuing training from global step 13000

[INFO|2025-02-22 11:27:23] trainer.py:2404 >>   Will skip the first 1 epochs then the first 1216 batches in the first epoch.

[INFO|2025-02-22 11:27:23] trainer.py:2643 >> 

Training completed. Do not forget to share your model on huggingface.co/models =)



[INFO|2025-02-22 11:27:23] trainer.py:3910 >> Saving model checkpoint to saves\Qwen2.5-0.5B-Instruct\full\20-02-2025

[INFO|2025-02-22 11:27:23] configuration_utils.py:420 >> Configuration saved in saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\config.json

[INFO|2025-02-22 11:27:23] configuration_utils.py:909 >> Configuration saved in saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\generation_config.json

[INFO|2025-02-22 11:27:25] modeling_utils.py:2988 >> Model weights saved in saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\model.safetensors

[INFO|2025-02-22 11:27:25] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\tokenizer_config.json

[INFO|2025-02-22 11:27:25] tokenization_utils_base.py:2500 >> Special tokens file saved in saves\Qwen2.5-0.5B-Instruct\full\20-02-2025\special_tokens_map.json

[WARNING|2025-02-22 11:27:26] logging.py:162 >> No metric eval_accuracy to plot.

[INFO|2025-02-22 11:27:26] trainer.py:4226 >> 
***** Running Evaluation *****

[INFO|2025-02-22 11:27:26] trainer.py:4228 >>   Num examples = 11421

[INFO|2025-02-22 11:27:26] trainer.py:4231 >>   Batch size = 1