File size: 11,264 Bytes
1d621f2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 |
05/05/2024 11:09:07 - INFO - transformers.tokenization_utils_base - loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/tokenizer.model
05/05/2024 11:09:07 - INFO - transformers.tokenization_utils_base - loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/tokenizer.json
05/05/2024 11:09:07 - INFO - transformers.tokenization_utils_base - loading file added_tokens.json from cache at None
05/05/2024 11:09:07 - INFO - transformers.tokenization_utils_base - loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/special_tokens_map.json
05/05/2024 11:09:07 - INFO - transformers.tokenization_utils_base - loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/tokenizer_config.json
05/05/2024 11:09:07 - INFO - llmtuner.data.loader - Loading dataset Reseacrhed_data.json...
05/05/2024 11:09:07 - WARNING - llmtuner.data.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
05/05/2024 11:09:09 - WARNING - transformers.tokenization_utils_base - Token indices sequence length is longer than the specified maximum sequence length for this model (2387 > 2048). Running this sequence through the model will result in indexing errors
05/05/2024 11:09:09 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/config.json
05/05/2024 11:09:09 - INFO - transformers.configuration_utils - Model config LlamaConfig {
"_name_or_path": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 5632,
"max_position_embeddings": 2048,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 22,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.1",
"use_cache": true,
"vocab_size": 32000
}
05/05/2024 11:09:09 - INFO - llmtuner.model.utils.quantization - Quantizing model to 8 bit.
05/05/2024 11:09:30 - INFO - transformers.modeling_utils - loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/model.safetensors
05/05/2024 11:09:30 - INFO - transformers.modeling_utils - Instantiating LlamaForCausalLM model under default dtype torch.float16.
05/05/2024 11:09:30 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
"bos_token_id": 1,
"eos_token_id": 2
}
05/05/2024 11:09:34 - INFO - transformers.modeling_utils - All model checkpoint weights were used when initializing LlamaForCausalLM.
05/05/2024 11:09:34 - INFO - transformers.modeling_utils - All the weights of LlamaForCausalLM were initialized from the model checkpoint at TinyLlama/TinyLlama-1.1B-Chat-v1.0.
If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
05/05/2024 11:09:35 - INFO - transformers.generation.configuration_utils - loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/generation_config.json
05/05/2024 11:09:35 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
"bos_token_id": 1,
"eos_token_id": 2,
"max_length": 2048,
"pad_token_id": 0
}
05/05/2024 11:09:35 - INFO - llmtuner.model.utils.checkpointing - Gradient checkpointing enabled.
05/05/2024 11:09:35 - INFO - llmtuner.model.utils.attention - Using torch SDPA for faster training and inference.
05/05/2024 11:09:35 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA
05/05/2024 11:09:35 - INFO - llmtuner.model.loader - trainable params: 1126400 || all params: 1101174784 || trainable%: 0.1023
05/05/2024 11:09:35 - INFO - transformers.trainer - Using auto half precision backend
05/05/2024 11:09:35 - INFO - transformers.trainer - ***** Running training *****
05/05/2024 11:09:35 - INFO - transformers.trainer - Num examples = 855
05/05/2024 11:09:35 - INFO - transformers.trainer - Num Epochs = 2
05/05/2024 11:09:35 - INFO - transformers.trainer - Instantaneous batch size per device = 2
05/05/2024 11:09:35 - INFO - transformers.trainer - Total train batch size (w. parallel, distributed & accumulation) = 16
05/05/2024 11:09:35 - INFO - transformers.trainer - Gradient Accumulation steps = 8
05/05/2024 11:09:35 - INFO - transformers.trainer - Total optimization steps = 106
05/05/2024 11:09:35 - INFO - transformers.trainer - Number of trainable parameters = 1,126,400
05/05/2024 11:10:23 - INFO - llmtuner.extras.callbacks - {'loss': 2.9430, 'learning_rate': 1.9890e-05, 'epoch': 0.09}
05/05/2024 11:11:11 - INFO - llmtuner.extras.callbacks - {'loss': 3.0203, 'learning_rate': 1.9564e-05, 'epoch': 0.19}
05/05/2024 11:11:59 - INFO - llmtuner.extras.callbacks - {'loss': 2.8306, 'learning_rate': 1.9028e-05, 'epoch': 0.28}
05/05/2024 11:12:48 - INFO - llmtuner.extras.callbacks - {'loss': 2.9983, 'learning_rate': 1.8294e-05, 'epoch': 0.37}
05/05/2024 11:13:37 - INFO - llmtuner.extras.callbacks - {'loss': 3.0059, 'learning_rate': 1.7378e-05, 'epoch': 0.47}
05/05/2024 11:14:26 - INFO - llmtuner.extras.callbacks - {'loss': 2.8450, 'learning_rate': 1.6301e-05, 'epoch': 0.56}
05/05/2024 11:15:16 - INFO - llmtuner.extras.callbacks - {'loss': 2.9230, 'learning_rate': 1.5085e-05, 'epoch': 0.65}
05/05/2024 11:16:05 - INFO - llmtuner.extras.callbacks - {'loss': 2.9725, 'learning_rate': 1.3758e-05, 'epoch': 0.75}
05/05/2024 11:16:55 - INFO - llmtuner.extras.callbacks - {'loss': 2.9861, 'learning_rate': 1.2349e-05, 'epoch': 0.84}
05/05/2024 11:17:45 - INFO - llmtuner.extras.callbacks - {'loss': 2.9897, 'learning_rate': 1.0888e-05, 'epoch': 0.93}
05/05/2024 11:18:35 - INFO - llmtuner.extras.callbacks - {'loss': 2.9677, 'learning_rate': 9.4076e-06, 'epoch': 1.03}
05/05/2024 11:19:24 - INFO - llmtuner.extras.callbacks - {'loss': 2.9620, 'learning_rate': 7.9402e-06, 'epoch': 1.12}
05/05/2024 11:20:14 - INFO - llmtuner.extras.callbacks - {'loss': 2.9335, 'learning_rate': 6.5180e-06, 'epoch': 1.21}
05/05/2024 11:21:04 - INFO - llmtuner.extras.callbacks - {'loss': 2.9682, 'learning_rate': 5.1721e-06, 'epoch': 1.31}
05/05/2024 11:21:54 - INFO - llmtuner.extras.callbacks - {'loss': 2.9475, 'learning_rate': 3.9320e-06, 'epoch': 1.40}
05/05/2024 11:22:43 - INFO - llmtuner.extras.callbacks - {'loss': 2.9373, 'learning_rate': 2.8249e-06, 'epoch': 1.50}
05/05/2024 11:23:34 - INFO - llmtuner.extras.callbacks - {'loss': 2.9115, 'learning_rate': 1.8751e-06, 'epoch': 1.59}
05/05/2024 11:24:24 - INFO - llmtuner.extras.callbacks - {'loss': 2.9237, 'learning_rate': 1.1034e-06, 'epoch': 1.68}
05/05/2024 11:25:13 - INFO - llmtuner.extras.callbacks - {'loss': 2.8901, 'learning_rate': 5.2674e-07, 'epoch': 1.78}
05/05/2024 11:26:03 - INFO - llmtuner.extras.callbacks - {'loss': 2.8905, 'learning_rate': 1.5769e-07, 'epoch': 1.87}
05/05/2024 11:26:03 - INFO - transformers.trainer - Saving model checkpoint to saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50/checkpoint-100
05/05/2024 11:26:03 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/config.json
05/05/2024 11:26:03 - INFO - transformers.configuration_utils - Model config LlamaConfig {
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 5632,
"max_position_embeddings": 2048,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 22,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.1",
"use_cache": true,
"vocab_size": 32000
}
05/05/2024 11:26:03 - INFO - transformers.tokenization_utils_base - tokenizer config file saved in saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50/checkpoint-100/tokenizer_config.json
05/05/2024 11:26:03 - INFO - transformers.tokenization_utils_base - Special tokens file saved in saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50/checkpoint-100/special_tokens_map.json
05/05/2024 11:26:53 - INFO - llmtuner.extras.callbacks - {'loss': 2.9548, 'learning_rate': 4.3916e-09, 'epoch': 1.96}
05/05/2024 11:27:03 - INFO - transformers.trainer -
Training completed. Do not forget to share your model on huggingface.co/models =)
05/05/2024 11:27:03 - INFO - transformers.trainer - Saving model checkpoint to saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50
05/05/2024 11:27:03 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--TinyLlama--TinyLlama-1.1B-Chat-v1.0/snapshots/fe8a4ea1ffedaf415f4da2f062534de366a451e6/config.json
05/05/2024 11:27:03 - INFO - transformers.configuration_utils - Model config LlamaConfig {
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 5632,
"max_position_embeddings": 2048,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 22,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.1",
"use_cache": true,
"vocab_size": 32000
}
05/05/2024 11:27:03 - INFO - transformers.tokenization_utils_base - tokenizer config file saved in saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50/tokenizer_config.json
05/05/2024 11:27:03 - INFO - transformers.tokenization_utils_base - Special tokens file saved in saves/LLaMA-tiny/lora/train_2024-05-05-10-57-50/special_tokens_map.json
05/05/2024 11:27:03 - INFO - transformers.modelcard - Dropping the following result as it does not have all the necessary fields:
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
|