[INFO|2025-07-07 19:00:02] configuration_utils.py:696 >> loading configuration file config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/config.json

[INFO|2025-07-07 19:00:02] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-Coder-7B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 3584,
  "initializer_range": 0.02,
  "intermediate_size": 18944,
  "max_position_embeddings": 32768,
  "max_window_layers": 28,
  "model_type": "qwen2",
  "num_attention_heads": 28,
  "num_hidden_layers": 28,
  "num_key_value_heads": 4,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.2",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}


[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file vocab.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/vocab.json

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file merges.txt from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/merges.txt

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file tokenizer.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/tokenizer.json

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file added_tokens.json from cache at None

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file special_tokens_map.json from cache at None

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file tokenizer_config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/tokenizer_config.json

[INFO|2025-07-07 19:00:02] tokenization_utils_base.py:2034 >> loading file chat_template.jinja from cache at None

[INFO|2025-07-07 19:00:03] 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|2025-07-07 19:00:03] configuration_utils.py:696 >> loading configuration file config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/config.json

[INFO|2025-07-07 19:00:03] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-Coder-7B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 3584,
  "initializer_range": 0.02,
  "intermediate_size": 18944,
  "max_position_embeddings": 32768,
  "max_window_layers": 28,
  "model_type": "qwen2",
  "num_attention_heads": 28,
  "num_hidden_layers": 28,
  "num_key_value_heads": 4,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.2",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}


[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file vocab.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/vocab.json

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file merges.txt from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/merges.txt

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file tokenizer.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/tokenizer.json

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file added_tokens.json from cache at None

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file special_tokens_map.json from cache at None

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file tokenizer_config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/tokenizer_config.json

[INFO|2025-07-07 19:00:04] tokenization_utils_base.py:2034 >> loading file chat_template.jinja from cache at None

[INFO|2025-07-07 19:00:04] 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|2025-07-07 19:00:04] logging.py:157 >> Add <|im_end|> to stop words.

[INFO|2025-07-07 19:00:04] logging.py:157 >> Loading dataset Codes3_query_filtered_553474_mark_less_than_8.0.json...

[INFO|2025-07-07 19:00:43] configuration_utils.py:696 >> loading configuration file config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/config.json

[INFO|2025-07-07 19:00:43] configuration_utils.py:768 >> Model config Qwen2Config {
  "_name_or_path": "Qwen/Qwen2.5-Coder-7B-Instruct",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 151643,
  "eos_token_id": 151645,
  "hidden_act": "silu",
  "hidden_size": 3584,
  "initializer_range": 0.02,
  "intermediate_size": 18944,
  "max_position_embeddings": 32768,
  "max_window_layers": 28,
  "model_type": "qwen2",
  "num_attention_heads": 28,
  "num_hidden_layers": 28,
  "num_key_value_heads": 4,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.48.2",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 152064
}


[WARNING|2025-07-07 19:00:43] logging.py:162 >> Input length is smaller than max length. Consider increase input length.

[INFO|2025-07-07 19:00:43] logging.py:157 >> Using llama3 scaling strategy and setting scaling factor to 1.0.

[INFO|2025-07-07 19:00:43] logging.py:157 >> Using block diagonal attention for sequence packing without cross-attention.

[INFO|2025-07-07 19:00:44] logging.py:157 >> Liger kernel has been applied to the model.

[INFO|2025-07-07 19:00:44] modeling_utils.py:3904 >> loading weights file model.safetensors from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/model.safetensors.index.json

[INFO|2025-07-07 19:00:44] modeling_utils.py:1582 >> Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.

[INFO|2025-07-07 19:00:44] configuration_utils.py:1140 >> Generate config GenerationConfig {
  "bos_token_id": 151643,
  "eos_token_id": 151645
}


[INFO|2025-07-07 19:00:52] modeling_utils.py:4888 >> All model checkpoint weights were used when initializing Qwen2ForCausalLM.


[INFO|2025-07-07 19:00:52] modeling_utils.py:4896 >> All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at Qwen/Qwen2.5-Coder-7B-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|2025-07-07 19:00:52] configuration_utils.py:1095 >> loading configuration file generation_config.json from cache at /home/kiho/.cache/huggingface/hub/models--Qwen--Qwen2.5-Coder-7B-Instruct/snapshots/c03e6d358207e414f1eca0bb1891e29f1db0e242/generation_config.json

[INFO|2025-07-07 19:00:52] 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|2025-07-07 19:00:52] logging.py:157 >> Gradient checkpointing enabled.

[INFO|2025-07-07 19:00:52] logging.py:157 >> Using torch SDPA for faster training and inference.

[INFO|2025-07-07 19:00:52] logging.py:157 >> Upcasting trainable params to float32.

[INFO|2025-07-07 19:00:52] logging.py:157 >> Fine-tuning method: Freeze

[INFO|2025-07-07 19:00:52] logging.py:157 >> Set trainable layers: .13.,.27.

[INFO|2025-07-07 19:00:52] logging.py:157 >> trainable params: 466,115,584 || all params: 7,615,616,512 || trainable%: 6.1205

[INFO|2025-07-07 19:00:52] trainer.py:741 >> Using auto half precision backend

[INFO|2025-07-07 19:00:53] logging.py:157 >> Found linear modules: q_proj,gate_proj,k_proj,down_proj,v_proj,o_proj,up_proj

[INFO|2025-07-07 19:00:53] logging.py:157 >> Using APOLLO optimizer with args: {'rank': 256, 'proj': 'random', 'proj_type': 'std', 'update_proj_gap': 200, 'scale': 1, 'scale_type': 'channel', 'scale_front': False}.

[INFO|2025-07-07 19:00:53] trainer.py:2369 >> ***** Running training *****

[INFO|2025-07-07 19:00:53] trainer.py:2370 >>   Num examples = 23,588

[INFO|2025-07-07 19:00:53] trainer.py:2371 >>   Num Epochs = 1

[INFO|2025-07-07 19:00:53] trainer.py:2372 >>   Instantaneous batch size per device = 16

[INFO|2025-07-07 19:00:53] trainer.py:2375 >>   Total train batch size (w. parallel, distributed & accumulation) = 384

[INFO|2025-07-07 19:00:53] trainer.py:2376 >>   Gradient Accumulation steps = 8

[INFO|2025-07-07 19:00:53] trainer.py:2377 >>   Total optimization steps = 61

[INFO|2025-07-07 19:00:53] trainer.py:2378 >>   Number of trainable parameters = 466,115,584

[INFO|2025-07-07 19:03:35] logging.py:157 >> {'loss': 0.8835, 'learning_rate': 4.9967e-05, 'epoch': 0.02, 'throughput': 9741.84}

[INFO|2025-07-07 19:06:09] logging.py:157 >> {'loss': 0.8172, 'learning_rate': 4.9867e-05, 'epoch': 0.03, 'throughput': 9978.43}

[INFO|2025-07-07 19:08:43] logging.py:157 >> {'loss': 0.7415, 'learning_rate': 4.9702e-05, 'epoch': 0.05, 'throughput': 10069.50}

[INFO|2025-07-07 19:11:16] logging.py:157 >> {'loss': 0.7198, 'learning_rate': 4.9471e-05, 'epoch': 0.07, 'throughput': 10117.11}

[INFO|2025-07-07 19:13:50] logging.py:157 >> {'loss': 0.6985, 'learning_rate': 4.9176e-05, 'epoch': 0.08, 'throughput': 10129.94}

[INFO|2025-07-07 19:16:24] logging.py:157 >> {'loss': 0.6642, 'learning_rate': 4.8816e-05, 'epoch': 0.10, 'throughput': 10146.98}

[INFO|2025-07-07 19:18:58] logging.py:157 >> {'loss': 0.6677, 'learning_rate': 4.8393e-05, 'epoch': 0.11, 'throughput': 10159.48}

[INFO|2025-07-07 19:21:32] logging.py:157 >> {'loss': 0.6451, 'learning_rate': 4.7908e-05, 'epoch': 0.13, 'throughput': 10164.48}

[INFO|2025-07-07 19:24:06] logging.py:157 >> {'loss': 0.6327, 'learning_rate': 4.7362e-05, 'epoch': 0.15, 'throughput': 10172.04}

[INFO|2025-07-07 19:26:41] logging.py:157 >> {'loss': 0.6331, 'learning_rate': 4.6757e-05, 'epoch': 0.16, 'throughput': 10169.75}

[INFO|2025-07-07 19:29:17] logging.py:157 >> {'loss': 0.6219, 'learning_rate': 4.6094e-05, 'epoch': 0.18, 'throughput': 10161.43}

[INFO|2025-07-07 19:31:50] logging.py:157 >> {'loss': 0.6205, 'learning_rate': 4.5376e-05, 'epoch': 0.20, 'throughput': 10168.14}

[INFO|2025-07-07 19:34:24] logging.py:157 >> {'loss': 0.6010, 'learning_rate': 4.4603e-05, 'epoch': 0.21, 'throughput': 10172.37}

[INFO|2025-07-07 19:36:58] logging.py:157 >> {'loss': 0.6077, 'learning_rate': 4.3778e-05, 'epoch': 0.23, 'throughput': 10175.33}

[INFO|2025-07-07 19:39:32] logging.py:157 >> {'loss': 0.5958, 'learning_rate': 4.2904e-05, 'epoch': 0.24, 'throughput': 10179.71}

[INFO|2025-07-07 19:42:05] logging.py:157 >> {'loss': 0.5838, 'learning_rate': 4.1982e-05, 'epoch': 0.26, 'throughput': 10183.00}

[INFO|2025-07-07 19:44:39] logging.py:157 >> {'loss': 0.5629, 'learning_rate': 4.1015e-05, 'epoch': 0.28, 'throughput': 10185.76}

[INFO|2025-07-07 19:47:13] logging.py:157 >> {'loss': 0.5848, 'learning_rate': 4.0005e-05, 'epoch': 0.29, 'throughput': 10188.71}

[INFO|2025-07-07 19:49:47] logging.py:157 >> {'loss': 0.5772, 'learning_rate': 3.8956e-05, 'epoch': 0.31, 'throughput': 10190.23}

[INFO|2025-07-07 19:52:20] logging.py:157 >> {'loss': 0.5719, 'learning_rate': 3.7870e-05, 'epoch': 0.33, 'throughput': 10192.39}

[INFO|2025-07-07 19:54:54] logging.py:157 >> {'loss': 0.5445, 'learning_rate': 3.6749e-05, 'epoch': 0.34, 'throughput': 10195.72}

[INFO|2025-07-07 19:57:27] logging.py:157 >> {'loss': 0.5560, 'learning_rate': 3.5598e-05, 'epoch': 0.36, 'throughput': 10198.63}

[INFO|2025-07-07 20:00:01] logging.py:157 >> {'loss': 0.5736, 'learning_rate': 3.4418e-05, 'epoch': 0.37, 'throughput': 10199.96}

[INFO|2025-07-07 20:02:34] logging.py:157 >> {'loss': 0.5350, 'learning_rate': 3.3214e-05, 'epoch': 0.39, 'throughput': 10203.32}

[INFO|2025-07-07 20:05:07] logging.py:157 >> {'loss': 0.5634, 'learning_rate': 3.1987e-05, 'epoch': 0.41, 'throughput': 10205.81}

[INFO|2025-07-07 20:07:42] logging.py:157 >> {'loss': 0.5648, 'learning_rate': 3.0742e-05, 'epoch': 0.42, 'throughput': 10203.48}

[INFO|2025-07-07 20:10:16] logging.py:157 >> {'loss': 0.5467, 'learning_rate': 2.9482e-05, 'epoch': 0.44, 'throughput': 10202.72}

[INFO|2025-07-07 20:12:50] logging.py:157 >> {'loss': 0.5567, 'learning_rate': 2.8210e-05, 'epoch': 0.46, 'throughput': 10203.80}

[INFO|2025-07-07 20:15:24] logging.py:157 >> {'loss': 0.5847, 'learning_rate': 2.6929e-05, 'epoch': 0.47, 'throughput': 10203.36}

[INFO|2025-07-07 20:17:58] logging.py:157 >> {'loss': 0.5429, 'learning_rate': 2.5644e-05, 'epoch': 0.49, 'throughput': 10204.19}

[INFO|2025-07-07 20:20:32] logging.py:157 >> {'loss': 0.5435, 'learning_rate': 2.4356e-05, 'epoch': 0.50, 'throughput': 10204.68}

[INFO|2025-07-07 20:23:06] logging.py:157 >> {'loss': 0.5482, 'learning_rate': 2.3071e-05, 'epoch': 0.52, 'throughput': 10205.83}

[INFO|2025-07-07 20:25:40] logging.py:157 >> {'loss': 0.5418, 'learning_rate': 2.1790e-05, 'epoch': 0.54, 'throughput': 10206.16}

[INFO|2025-07-07 20:28:14] logging.py:157 >> {'loss': 0.5320, 'learning_rate': 2.0518e-05, 'epoch': 0.55, 'throughput': 10205.44}

[INFO|2025-07-07 20:30:48] logging.py:157 >> {'loss': 0.5360, 'learning_rate': 1.9258e-05, 'epoch': 0.57, 'throughput': 10205.42}

[INFO|2025-07-07 20:33:23] logging.py:157 >> {'loss': 0.5314, 'learning_rate': 1.8013e-05, 'epoch': 0.59, 'throughput': 10204.05}

[INFO|2025-07-07 20:35:56] logging.py:157 >> {'loss': 0.5595, 'learning_rate': 1.6786e-05, 'epoch': 0.60, 'throughput': 10205.77}

[INFO|2025-07-07 20:38:32] logging.py:157 >> {'loss': 0.5418, 'learning_rate': 1.5582e-05, 'epoch': 0.62, 'throughput': 10203.12}

[INFO|2025-07-07 20:41:06] logging.py:157 >> {'loss': 0.5438, 'learning_rate': 1.4402e-05, 'epoch': 0.63, 'throughput': 10203.74}

[INFO|2025-07-07 20:43:39] logging.py:157 >> {'loss': 0.5239, 'learning_rate': 1.3251e-05, 'epoch': 0.65, 'throughput': 10204.70}

[INFO|2025-07-07 20:46:13] logging.py:157 >> {'loss': 0.5459, 'learning_rate': 1.2130e-05, 'epoch': 0.67, 'throughput': 10204.78}

[INFO|2025-07-07 20:48:47] logging.py:157 >> {'loss': 0.5373, 'learning_rate': 1.1044e-05, 'epoch': 0.68, 'throughput': 10204.67}

[INFO|2025-07-07 20:51:22] logging.py:157 >> {'loss': 0.5474, 'learning_rate': 9.9946e-06, 'epoch': 0.70, 'throughput': 10203.24}

[INFO|2025-07-07 20:53:56] logging.py:157 >> {'loss': 0.5236, 'learning_rate': 8.9852e-06, 'epoch': 0.72, 'throughput': 10203.82}

[INFO|2025-07-07 20:56:29] logging.py:157 >> {'loss': 0.5336, 'learning_rate': 8.0182e-06, 'epoch': 0.73, 'throughput': 10205.18}

[INFO|2025-07-07 20:59:05] logging.py:157 >> {'loss': 0.5198, 'learning_rate': 7.0962e-06, 'epoch': 0.75, 'throughput': 10203.27}

[INFO|2025-07-07 21:01:40] logging.py:157 >> {'loss': 0.5419, 'learning_rate': 6.2217e-06, 'epoch': 0.76, 'throughput': 10202.64}

[INFO|2025-07-07 21:04:17] logging.py:157 >> {'loss': 0.5544, 'learning_rate': 5.3970e-06, 'epoch': 0.78, 'throughput': 10198.78}

[INFO|2025-07-07 21:06:50] logging.py:157 >> {'loss': 0.5689, 'learning_rate': 4.6243e-06, 'epoch': 0.80, 'throughput': 10200.12}

[INFO|2025-07-07 21:09:24] logging.py:157 >> {'loss': 0.5170, 'learning_rate': 3.9056e-06, 'epoch': 0.81, 'throughput': 10200.21}

[INFO|2025-07-07 21:11:58] logging.py:157 >> {'loss': 0.5440, 'learning_rate': 3.2429e-06, 'epoch': 0.83, 'throughput': 10200.78}

[INFO|2025-07-07 21:14:34] logging.py:157 >> {'loss': 0.5265, 'learning_rate': 2.6378e-06, 'epoch': 0.85, 'throughput': 10198.05}

[INFO|2025-07-07 21:17:09] logging.py:157 >> {'loss': 0.5301, 'learning_rate': 2.0921e-06, 'epoch': 0.86, 'throughput': 10197.60}

[INFO|2025-07-07 21:19:44] logging.py:157 >> {'loss': 0.5357, 'learning_rate': 1.6071e-06, 'epoch': 0.88, 'throughput': 10196.24}

[INFO|2025-07-07 21:22:18] logging.py:157 >> {'loss': 0.5220, 'learning_rate': 1.1841e-06, 'epoch': 0.89, 'throughput': 10196.15}

[INFO|2025-07-07 21:24:54] logging.py:157 >> {'loss': 0.5282, 'learning_rate': 8.2431e-07, 'epoch': 0.91, 'throughput': 10194.76}

[INFO|2025-07-07 21:27:28] logging.py:157 >> {'loss': 0.5327, 'learning_rate': 5.2861e-07, 'epoch': 0.93, 'throughput': 10195.16}

[INFO|2025-07-07 21:30:02] logging.py:157 >> {'loss': 0.5237, 'learning_rate': 2.9780e-07, 'epoch': 0.94, 'throughput': 10195.52}

[INFO|2025-07-07 21:32:36] logging.py:157 >> {'loss': 0.5499, 'learning_rate': 1.3250e-07, 'epoch': 0.96, 'throughput': 10195.79}

[INFO|2025-07-07 21:35:10] logging.py:157 >> {'loss': 0.5576, 'learning_rate': 3.3148e-08, 'epoch': 0.98, 'throughput': 10195.29}

[INFO|2025-07-07 21:37:45] logging.py:157 >> {'loss': 0.5487, 'learning_rate': 0.0000e+00, 'epoch': 0.99, 'throughput': 10195.34}

[INFO|2025-07-07 21:37:45] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61

[INFO|2025-07-07 21:37:45] configuration_utils.py:420 >> Configuration saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61/config.json

[INFO|2025-07-07 21:37:45] configuration_utils.py:909 >> Configuration saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61/generation_config.json

[INFO|2025-07-07 21:38:10] modeling_utils.py:2996 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61/model.safetensors.index.json.

[INFO|2025-07-07 21:38:10] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61/tokenizer_config.json

[INFO|2025-07-07 21:38:10] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/checkpoint-61/special_tokens_map.json

[INFO|2025-07-07 21:38:10] trainer.py:2643 >> 

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



[INFO|2025-07-07 21:38:10] trainer.py:3910 >> Saving model checkpoint to saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx

[INFO|2025-07-07 21:38:10] configuration_utils.py:420 >> Configuration saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/config.json

[INFO|2025-07-07 21:38:10] configuration_utils.py:909 >> Configuration saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/generation_config.json

[INFO|2025-07-07 21:38:35] modeling_utils.py:2996 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/model.safetensors.index.json.

[INFO|2025-07-07 21:38:35] tokenization_utils_base.py:2491 >> tokenizer config file saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/tokenizer_config.json

[INFO|2025-07-07 21:38:35] tokenization_utils_base.py:2500 >> Special tokens file saved in saves/Qwen2.5-Coder-7B-Instruct/freeze/qwen_under8_nlx/special_tokens_map.json

[WARNING|2025-07-07 21:38:35] logging.py:162 >> No metric eval_loss to plot.

[WARNING|2025-07-07 21:38:35] logging.py:162 >> No metric eval_accuracy to plot.

[INFO|2025-07-07 21:38:35] modelcard.py:449 >> Dropping the following result as it does not have all the necessary fields:
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}