Hugging Face's logo

Terminator-of-AI
/
Qwen1.5-0.5B-finetuning-by-super-battleship

Text Generation
Transformers
Safetensors
qwen2
llama-factory
full
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
lyc123456 commited on
Commit
7969749
·
verified ·
1 Parent(s): a8f12f7

upload model

Browse files
Files changed (14) hide show
  1. README.md +64 -1
  2. added_tokens.json +5 -0
  3. all_results.json +7 -0
  4. config.json +28 -0
  5. generation_config.json +6 -0
  6. main.py +45 -0
  7. merges.txt +0 -0
  8. special_tokens_map.json +20 -0
  9. tokenizer_config.json +44 -0
  10. train_results.json +7 -0
  11. trainer_log.jsonl +123 -0
  12. trainer_state.json +884 -0
  13. training_args.bin +3 -0
  14. vocab.json +0 -0
README.md CHANGED
@@ -1,3 +1,66 @@
1
  ---
2
- license: mit
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: other
3
+ base_model: Qwen1.5-0.5B
4
+ tags:
5
+ - llama-factory
6
+ - full
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: train_2024-03-31-14-36-00-superzj
10
+ results: []
11
  ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # train_2024-03-31-14-36-00-superzj
17
+
18
+ This model is a fine-tuned version of [Qwen1.5-0.5B](https://huggingface.co/Qwen/Qwen1.5-0.5B) on the superzj dataset.
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
+ - train_batch_size: 4
39
+ - eval_batch_size: 8
40
+ - seed: 42
41
+ - gradient_accumulation_steps: 8
42
+ - total_train_batch_size: 32
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: cosine
45
+ - num_epochs: 10.0
46
+
47
+ ### Training results
48
+
49
+
50
+
51
+ ### Framework versions
52
+
53
+ - Transformers 4.38.2
54
+
55
+ - Pytorch 1.13.1+cu116
56
+
57
+ - Datasets 2.14.6
58
+
59
+ - Tokenizers 0.15.2
60
+
61
+ ```
62
+ 全量微调，使用语料来自小说--重生之超级战舰
63
+ Fully fine-tuned, using corpus from the novel - Rebirth of the Super Battleship
64
+ ```
65
+
66
+
added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 1.2541689154554585,
4
+ "train_runtime": 15808.2669,
5
+ "train_samples_per_second": 1.233,
6
+ "train_steps_per_second": 0.039
7
+ }
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen1.5-0.5B-finetuning-by-super-battleship",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1024,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 2816,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 21,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 16,
17
+ "num_hidden_layers": 24,
18
+ "num_key_value_heads": 16,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_theta": 1000000.0,
21
+ "sliding_window": 32768,
22
+ "tie_word_embeddings": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.38.2",
25
+ "use_cache": false,
26
+ "use_sliding_window": false,
27
+ "vocab_size": 151936
28
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.38.2"
6
+ }
main.py ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ '''
2
+ Calling example, for reference only
3
+ 调用示例，仅供参考
4
+ '''
5
+ import os
6
+
7
+ from transformers import AutoModelForCausalLM, AutoTokenizer
8
+
9
+ messages = [
10
+ {"role": "system", "content": "You are a helpful assistant."},
11
+ ]
12
+
13
+ device = "cuda" # the device to load the model onto
14
+ model_path = os.path.dirname(__file__)
15
+ model = AutoModelForCausalLM.from_pretrained(
16
+ model_path,
17
+ torch_dtype="auto",
18
+ device_map="auto"
19
+ )
20
+ tokenizer = AutoTokenizer.from_pretrained(model_path)
21
+ response = ''
22
+ if __name__ == '__main__':
23
+
24
+ while True:
25
+ # prompt = "Give me a short introduction to large language model."
26
+ prompt = input("input:")
27
+ messages.append({"role": "user", "content": prompt})
28
+ text = tokenizer.apply_chat_template(
29
+ messages,
30
+ tokenize=False,
31
+ add_generation_prompt=True
32
+ )
33
+ model_inputs = tokenizer([text], return_tensors="pt").to(device)
34
+
35
+ generated_ids = model.generate(
36
+ model_inputs.input_ids,
37
+ max_new_tokens=512
38
+ )
39
+ generated_ids = [
40
+ output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
41
+ ]
42
+
43
+ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
44
+ print(response)
45
+ messages.append({"role": "system", "content": response}, )
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "eos_token": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "pad_token": {
14
+ "content": "<|endoftext|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ }
20
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ }
28
+ },
29
+ "additional_special_tokens": [
30
+ "<|im_start|>",
31
+ "<|im_end|>"
32
+ ],
33
+ "bos_token": null,
34
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '<|endoftext|>' + '\\n' }}{% endif %}{% endfor %}",
35
+ "clean_up_tokenization_spaces": false,
36
+ "eos_token": "<|endoftext|>",
37
+ "errors": "replace",
38
+ "model_max_length": 32768,
39
+ "pad_token": "<|endoftext|>",
40
+ "padding_side": "right",
41
+ "split_special_tokens": false,
42
+ "tokenizer_class": "Qwen2Tokenizer",
43
+ "unk_token": null
44
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 1.2541689154554585,
4
+ "train_runtime": 15808.2669,
5
+ "train_samples_per_second": 1.233,
6
+ "train_steps_per_second": 0.039
7
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 5, "total_steps": 610, "loss": 3.4699, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9991711687857826e-05, "epoch": 0.08, "percentage": 0.82, "elapsed_time": "0:02:12", "remaining_time": "4:28:10"}
2
+ {"current_steps": 10, "total_steps": 610, "loss": 3.1968, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9966852247120764e-05, "epoch": 0.16, "percentage": 1.64, "elapsed_time": "0:04:26", "remaining_time": "4:26:26"}
3
+ {"current_steps": 15, "total_steps": 610, "loss": 3.0675, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9925438161213164e-05, "epoch": 0.25, "percentage": 2.46, "elapsed_time": "0:06:52", "remaining_time": "4:32:24"}
4
+ {"current_steps": 20, "total_steps": 610, "loss": 3.0402, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9867496890364726e-05, "epoch": 0.33, "percentage": 3.28, "elapsed_time": "0:09:00", "remaining_time": "4:25:56"}
5
+ {"current_steps": 25, "total_steps": 610, "loss": 3.0463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9793066853402536e-05, "epoch": 0.41, "percentage": 4.1, "elapsed_time": "0:11:09", "remaining_time": "4:21:10"}
6
+ {"current_steps": 30, "total_steps": 610, "loss": 2.9922, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.970219740227693e-05, "epoch": 0.49, "percentage": 4.92, "elapsed_time": "0:13:18", "remaining_time": "4:17:17"}
7
+ {"current_steps": 35, "total_steps": 610, "loss": 2.9401, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9594948789337914e-05, "epoch": 0.57, "percentage": 5.74, "elapsed_time": "0:15:27", "remaining_time": "4:13:54"}
8
+ {"current_steps": 40, "total_steps": 610, "loss": 2.913, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.947139212738395e-05, "epoch": 0.66, "percentage": 6.56, "elapsed_time": "0:17:36", "remaining_time": "4:10:50"}
9
+ {"current_steps": 45, "total_steps": 610, "loss": 2.9039, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.933160934250957e-05, "epoch": 0.74, "percentage": 7.38, "elapsed_time": "0:19:45", "remaining_time": "4:07:59"}
10
+ {"current_steps": 50, "total_steps": 610, "loss": 2.9188, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9175693119783013e-05, "epoch": 0.82, "percentage": 8.2, "elapsed_time": "0:21:53", "remaining_time": "4:05:15"}
11
+ {"current_steps": 55, "total_steps": 610, "loss": 2.8791, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.900374684179004e-05, "epoch": 0.9, "percentage": 9.02, "elapsed_time": "0:24:02", "remaining_time": "4:02:39"}
12
+ {"current_steps": 60, "total_steps": 610, "loss": 2.8602, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.881588452008456e-05, "epoch": 0.98, "percentage": 9.84, "elapsed_time": "0:26:11", "remaining_time": "4:00:07"}
13
+ {"current_steps": 65, "total_steps": 610, "loss": 2.6114, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.861223071959153e-05, "epoch": 1.07, "percentage": 10.66, "elapsed_time": "0:28:18", "remaining_time": "3:57:19"}
14
+ {"current_steps": 70, "total_steps": 610, "loss": 2.5097, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.839292047601234e-05, "epoch": 1.15, "percentage": 11.48, "elapsed_time": "0:30:27", "remaining_time": "3:54:55"}
15
+ {"current_steps": 75, "total_steps": 610, "loss": 2.497, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.815809920628738e-05, "epoch": 1.23, "percentage": 12.3, "elapsed_time": "0:32:36", "remaining_time": "3:52:33"}
16
+ {"current_steps": 80, "total_steps": 610, "loss": 2.4959, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.790792261217512e-05, "epoch": 1.31, "percentage": 13.11, "elapsed_time": "0:34:45", "remaining_time": "3:50:13"}
17
+ {"current_steps": 85, "total_steps": 610, "loss": 2.5037, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.764255657701179e-05, "epoch": 1.39, "percentage": 13.93, "elapsed_time": "0:36:53", "remaining_time": "3:47:54"}
18
+ {"current_steps": 90, "total_steps": 610, "loss": 2.4973, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.736217705571989e-05, "epoch": 1.48, "percentage": 14.75, "elapsed_time": "0:39:02", "remaining_time": "3:45:37"}
19
+ {"current_steps": 95, "total_steps": 610, "loss": 2.4948, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.706696995813868e-05, "epoch": 1.56, "percentage": 15.57, "elapsed_time": "0:41:12", "remaining_time": "3:43:21"}
20
+ {"current_steps": 100, "total_steps": 610, "loss": 2.5013, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6757131025753886e-05, "epoch": 1.64, "percentage": 16.39, "elapsed_time": "0:43:21", "remaining_time": "3:41:05"}
21
+ {"current_steps": 105, "total_steps": 610, "loss": 2.4919, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.643286570190831e-05, "epoch": 1.72, "percentage": 17.21, "elapsed_time": "0:45:40", "remaining_time": "3:39:38"}
22
+ {"current_steps": 110, "total_steps": 610, "loss": 2.4871, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.609438899557964e-05, "epoch": 1.8, "percentage": 18.03, "elapsed_time": "0:47:49", "remaining_time": "3:37:21"}
23
+ {"current_steps": 115, "total_steps": 610, "loss": 2.5035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5741925338815474e-05, "epoch": 1.89, "percentage": 18.85, "elapsed_time": "0:49:58", "remaining_time": "3:35:04"}
24
+ {"current_steps": 120, "total_steps": 610, "loss": 2.4941, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5375708437920284e-05, "epoch": 1.97, "percentage": 19.67, "elapsed_time": "0:52:07", "remaining_time": "3:32:49"}
25
+ {"current_steps": 125, "total_steps": 610, "loss": 2.2503, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.499598111849299e-05, "epoch": 2.05, "percentage": 20.49, "elapsed_time": "0:54:14", "remaining_time": "3:30:25"}
26
+ {"current_steps": 130, "total_steps": 610, "loss": 2.0722, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.460299516441776e-05, "epoch": 2.13, "percentage": 21.31, "elapsed_time": "0:56:23", "remaining_time": "3:28:11"}
27
+ {"current_steps": 135, "total_steps": 610, "loss": 2.0567, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4197011150915e-05, "epoch": 2.21, "percentage": 22.13, "elapsed_time": "0:58:32", "remaining_time": "3:25:57"}
28
+ {"current_steps": 140, "total_steps": 610, "loss": 2.0507, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3778298271762995e-05, "epoch": 2.3, "percentage": 22.95, "elapsed_time": "1:00:41", "remaining_time": "3:23:44"}
29
+ {"current_steps": 145, "total_steps": 610, "loss": 2.0694, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.334713416080498e-05, "epoch": 2.38, "percentage": 23.77, "elapsed_time": "1:02:50", "remaining_time": "3:21:31"}
30
+ {"current_steps": 150, "total_steps": 610, "loss": 2.0711, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2903804707859835e-05, "epoch": 2.46, "percentage": 24.59, "elapsed_time": "1:04:59", "remaining_time": "3:19:18"}
31
+ {"current_steps": 155, "total_steps": 610, "loss": 2.0782, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2448603869158587e-05, "epoch": 2.54, "percentage": 25.41, "elapsed_time": "1:07:08", "remaining_time": "3:17:05"}
32
+ {"current_steps": 160, "total_steps": 610, "loss": 2.0982, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.198183347243233e-05, "epoch": 2.62, "percentage": 26.23, "elapsed_time": "1:09:17", "remaining_time": "3:14:53"}
33
+ {"current_steps": 165, "total_steps": 610, "loss": 2.1043, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1503803016780796e-05, "epoch": 2.7, "percentage": 27.05, "elapsed_time": "1:11:26", "remaining_time": "3:12:41"}
34
+ {"current_steps": 170, "total_steps": 610, "loss": 2.0925, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.101482946745439e-05, "epoch": 2.79, "percentage": 27.87, "elapsed_time": "1:13:35", "remaining_time": "3:10:29"}
35
+ {"current_steps": 175, "total_steps": 610, "loss": 2.1157, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.051523704568557e-05, "epoch": 2.87, "percentage": 28.69, "elapsed_time": "1:15:45", "remaining_time": "3:08:17"}
36
+ {"current_steps": 180, "total_steps": 610, "loss": 2.1207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.000535701370921e-05, "epoch": 2.95, "percentage": 29.51, "elapsed_time": "1:17:54", "remaining_time": "3:06:06"}
37
+ {"current_steps": 185, "total_steps": 610, "loss": 1.9371, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.948552745511409e-05, "epoch": 3.03, "percentage": 30.33, "elapsed_time": "1:20:01", "remaining_time": "3:03:49"}
38
+ {"current_steps": 190, "total_steps": 610, "loss": 1.6614, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.895609305067162e-05, "epoch": 3.11, "percentage": 31.15, "elapsed_time": "1:22:10", "remaining_time": "3:01:38"}
39
+ {"current_steps": 195, "total_steps": 610, "loss": 1.6199, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.841740484979002e-05, "epoch": 3.2, "percentage": 31.97, "elapsed_time": "1:24:19", "remaining_time": "2:59:27"}
40
+ {"current_steps": 200, "total_steps": 610, "loss": 1.6182, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.7869820037745776e-05, "epoch": 3.28, "percentage": 32.79, "elapsed_time": "1:26:28", "remaining_time": "2:57:16"}
41
+ {"current_steps": 205, "total_steps": 610, "loss": 1.613, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.731370169884662e-05, "epoch": 3.36, "percentage": 33.61, "elapsed_time": "1:28:47", "remaining_time": "2:55:24"}
42
+ {"current_steps": 210, "total_steps": 610, "loss": 1.6146, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6749418575683e-05, "epoch": 3.44, "percentage": 34.43, "elapsed_time": "1:30:56", "remaining_time": "2:53:13"}
43
+ {"current_steps": 215, "total_steps": 610, "loss": 1.6147, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.617734482462785e-05, "epoch": 3.52, "percentage": 35.25, "elapsed_time": "1:33:05", "remaining_time": "2:51:01"}
44
+ {"current_steps": 220, "total_steps": 610, "loss": 1.6139, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5597859767746524e-05, "epoch": 3.61, "percentage": 36.07, "elapsed_time": "1:35:14", "remaining_time": "2:48:50"}
45
+ {"current_steps": 225, "total_steps": 610, "loss": 1.6382, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.501134764128167e-05, "epoch": 3.69, "percentage": 36.89, "elapsed_time": "1:37:23", "remaining_time": "2:46:39"}
46
+ {"current_steps": 230, "total_steps": 610, "loss": 1.6468, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4418197340879635e-05, "epoch": 3.77, "percentage": 37.7, "elapsed_time": "1:39:32", "remaining_time": "2:44:27"}
47
+ {"current_steps": 235, "total_steps": 610, "loss": 1.6518, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.381880216372738e-05, "epoch": 3.85, "percentage": 38.52, "elapsed_time": "1:41:41", "remaining_time": "2:42:16"}
48
+ {"current_steps": 240, "total_steps": 610, "loss": 1.6574, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.321355954777087e-05, "epoch": 3.93, "percentage": 39.34, "elapsed_time": "1:43:50", "remaining_time": "2:40:05"}
49
+ {"current_steps": 245, "total_steps": 610, "loss": 1.5766, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.260287080818795e-05, "epoch": 4.02, "percentage": 40.16, "elapsed_time": "1:45:57", "remaining_time": "2:37:51"}
50
+ {"current_steps": 250, "total_steps": 610, "loss": 1.2207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1987140871290236e-05, "epoch": 4.1, "percentage": 40.98, "elapsed_time": "1:48:07", "remaining_time": "2:35:41"}
51
+ {"current_steps": 255, "total_steps": 610, "loss": 1.202, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.136677800603072e-05, "epoch": 4.18, "percentage": 41.8, "elapsed_time": "1:50:16", "remaining_time": "2:33:30"}
52
+ {"current_steps": 260, "total_steps": 610, "loss": 1.1862, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.07421935532949e-05, "epoch": 4.26, "percentage": 42.62, "elapsed_time": "1:52:25", "remaining_time": "2:31:20"}
53
+ {"current_steps": 265, "total_steps": 610, "loss": 1.1969, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0113801653155026e-05, "epoch": 4.34, "percentage": 43.44, "elapsed_time": "1:54:34", "remaining_time": "2:29:09"}
54
+ {"current_steps": 270, "total_steps": 610, "loss": 1.1731, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9482018970268393e-05, "epoch": 4.43, "percentage": 44.26, "elapsed_time": "1:56:43", "remaining_time": "2:26:59"}
55
+ {"current_steps": 275, "total_steps": 610, "loss": 1.1747, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.884726441760155e-05, "epoch": 4.51, "percentage": 45.08, "elapsed_time": "1:58:52", "remaining_time": "2:24:48"}
56
+ {"current_steps": 280, "total_steps": 610, "loss": 1.1924, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8209958878663778e-05, "epoch": 4.59, "percentage": 45.9, "elapsed_time": "2:01:01", "remaining_time": "2:22:38"}
57
+ {"current_steps": 285, "total_steps": 610, "loss": 1.1844, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.757052492843401e-05, "epoch": 4.67, "percentage": 46.72, "elapsed_time": "2:03:11", "remaining_time": "2:20:28"}
58
+ {"current_steps": 290, "total_steps": 610, "loss": 1.1892, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6929386553166164e-05, "epoch": 4.75, "percentage": 47.54, "elapsed_time": "2:05:20", "remaining_time": "2:18:18"}
59
+ {"current_steps": 295, "total_steps": 610, "loss": 1.2005, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6286968869258665e-05, "epoch": 4.84, "percentage": 48.36, "elapsed_time": "2:07:29", "remaining_time": "2:16:08"}
60
+ {"current_steps": 300, "total_steps": 610, "loss": 1.2167, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.564369784137472e-05, "epoch": 4.92, "percentage": 49.18, "elapsed_time": "2:09:38", "remaining_time": "2:13:58"}
61
+ {"current_steps": 305, "total_steps": 610, "loss": 1.2292, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5e-05, "epoch": 5.0, "percentage": 50.0, "elapsed_time": "2:11:54", "remaining_time": "2:11:54"}
62
+ {"current_steps": 310, "total_steps": 610, "loss": 0.8431, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4356302158625288e-05, "epoch": 5.08, "percentage": 50.82, "elapsed_time": "2:14:04", "remaining_time": "2:09:44"}
63
+ {"current_steps": 315, "total_steps": 610, "loss": 0.8032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3713031130741337e-05, "epoch": 5.16, "percentage": 51.64, "elapsed_time": "2:16:13", "remaining_time": "2:07:34"}
64
+ {"current_steps": 320, "total_steps": 610, "loss": 0.8136, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3070613446833842e-05, "epoch": 5.25, "percentage": 52.46, "elapsed_time": "2:18:22", "remaining_time": "2:05:24"}
65
+ {"current_steps": 325, "total_steps": 610, "loss": 0.813, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2429475071565987e-05, "epoch": 5.33, "percentage": 53.28, "elapsed_time": "2:20:31", "remaining_time": "2:03:14"}
66
+ {"current_steps": 330, "total_steps": 610, "loss": 0.8105, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1790041121336225e-05, "epoch": 5.41, "percentage": 54.1, "elapsed_time": "2:22:41", "remaining_time": "2:01:03"}
67
+ {"current_steps": 335, "total_steps": 610, "loss": 0.8095, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1152735582398452e-05, "epoch": 5.49, "percentage": 54.92, "elapsed_time": "2:24:50", "remaining_time": "1:58:53"}
68
+ {"current_steps": 340, "total_steps": 610, "loss": 0.8083, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0517981029731616e-05, "epoch": 5.57, "percentage": 55.74, "elapsed_time": "2:26:59", "remaining_time": "1:56:43"}
69
+ {"current_steps": 345, "total_steps": 610, "loss": 0.8329, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9886198346844987e-05, "epoch": 5.66, "percentage": 56.56, "elapsed_time": "2:29:08", "remaining_time": "1:54:33"}
70
+ {"current_steps": 350, "total_steps": 610, "loss": 0.8162, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9257806446705116e-05, "epoch": 5.74, "percentage": 57.38, "elapsed_time": "2:31:17", "remaining_time": "1:52:23"}
71
+ {"current_steps": 355, "total_steps": 610, "loss": 0.8286, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8633221993969285e-05, "epoch": 5.82, "percentage": 58.2, "elapsed_time": "2:33:26", "remaining_time": "1:50:13"}
72
+ {"current_steps": 360, "total_steps": 610, "loss": 0.8318, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8012859128709766e-05, "epoch": 5.9, "percentage": 59.02, "elapsed_time": "2:35:35", "remaining_time": "1:48:03"}
73
+ {"current_steps": 365, "total_steps": 610, "loss": 0.8312, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7397129191812057e-05, "epoch": 5.98, "percentage": 59.84, "elapsed_time": "2:37:45", "remaining_time": "1:45:53"}
74
+ {"current_steps": 370, "total_steps": 610, "loss": 0.6165, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6786440452229134e-05, "epoch": 6.07, "percentage": 60.66, "elapsed_time": "2:39:51", "remaining_time": "1:43:41"}
75
+ {"current_steps": 375, "total_steps": 610, "loss": 0.5368, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.618119783627263e-05, "epoch": 6.15, "percentage": 61.48, "elapsed_time": "2:42:01", "remaining_time": "1:41:31"}
76
+ {"current_steps": 380, "total_steps": 610, "loss": 0.5347, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.558180265912037e-05, "epoch": 6.23, "percentage": 62.3, "elapsed_time": "2:44:10", "remaining_time": "1:39:21"}
77
+ {"current_steps": 385, "total_steps": 610, "loss": 0.5234, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4988652358718336e-05, "epoch": 6.31, "percentage": 63.11, "elapsed_time": "2:46:19", "remaining_time": "1:37:12"}
78
+ {"current_steps": 390, "total_steps": 610, "loss": 0.514, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4402140232253486e-05, "epoch": 6.39, "percentage": 63.93, "elapsed_time": "2:48:28", "remaining_time": "1:35:02"}
79
+ {"current_steps": 395, "total_steps": 610, "loss": 0.522, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3822655175372149e-05, "epoch": 6.48, "percentage": 64.75, "elapsed_time": "2:50:37", "remaining_time": "1:32:52"}
80
+ {"current_steps": 400, "total_steps": 610, "loss": 0.5207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.325058142431701e-05, "epoch": 6.56, "percentage": 65.57, "elapsed_time": "2:52:46", "remaining_time": "1:30:42"}
81
+ {"current_steps": 405, "total_steps": 610, "loss": 0.5399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2686298301153393e-05, "epoch": 6.64, "percentage": 66.39, "elapsed_time": "2:55:05", "remaining_time": "1:28:37"}
82
+ {"current_steps": 410, "total_steps": 610, "loss": 0.5304, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.213017996225424e-05, "epoch": 6.72, "percentage": 67.21, "elapsed_time": "2:57:14", "remaining_time": "1:26:27"}
83
+ {"current_steps": 415, "total_steps": 610, "loss": 0.534, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.158259515020999e-05, "epoch": 6.8, "percentage": 68.03, "elapsed_time": "2:59:23", "remaining_time": "1:24:17"}
84
+ {"current_steps": 420, "total_steps": 610, "loss": 0.5255, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1043906949328387e-05, "epoch": 6.89, "percentage": 68.85, "elapsed_time": "3:01:32", "remaining_time": "1:22:07"}
85
+ {"current_steps": 425, "total_steps": 610, "loss": 0.5388, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.051447254488591e-05, "epoch": 6.97, "percentage": 69.67, "elapsed_time": "3:03:42", "remaining_time": "1:19:57"}
86
+ {"current_steps": 430, "total_steps": 610, "loss": 0.4218, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.9946429862908e-06, "epoch": 7.05, "percentage": 70.49, "elapsed_time": "3:05:49", "remaining_time": "1:17:47"}
87
+ {"current_steps": 435, "total_steps": 610, "loss": 0.3403, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.48476295431443e-06, "epoch": 7.13, "percentage": 71.31, "elapsed_time": "3:07:58", "remaining_time": "1:15:37"}
88
+ {"current_steps": 440, "total_steps": 610, "loss": 0.3332, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.985170532545622e-06, "epoch": 7.21, "percentage": 72.13, "elapsed_time": "3:10:07", "remaining_time": "1:13:27"}
89
+ {"current_steps": 445, "total_steps": 610, "loss": 0.3341, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.496196983219203e-06, "epoch": 7.3, "percentage": 72.95, "elapsed_time": "3:12:16", "remaining_time": "1:11:17"}
90
+ {"current_steps": 450, "total_steps": 610, "loss": 0.3334, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.018166527567672e-06, "epoch": 7.38, "percentage": 73.77, "elapsed_time": "3:14:25", "remaining_time": "1:09:07"}
91
+ {"current_steps": 455, "total_steps": 610, "loss": 0.3329, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.5513961308414065e-06, "epoch": 7.46, "percentage": 74.59, "elapsed_time": "3:16:35", "remaining_time": "1:06:58"}
92
+ {"current_steps": 460, "total_steps": 610, "loss": 0.3407, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.096195292140173e-06, "epoch": 7.54, "percentage": 75.41, "elapsed_time": "3:18:44", "remaining_time": "1:04:48"}
93
+ {"current_steps": 465, "total_steps": 610, "loss": 0.3329, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.652865839195024e-06, "epoch": 7.62, "percentage": 76.23, "elapsed_time": "3:20:53", "remaining_time": "1:02:38"}
94
+ {"current_steps": 470, "total_steps": 610, "loss": 0.3314, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.221701728237009e-06, "epoch": 7.7, "percentage": 77.05, "elapsed_time": "3:23:02", "remaining_time": "1:00:28"}
95
+ {"current_steps": 475, "total_steps": 610, "loss": 0.3388, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.8029888490850005e-06, "epoch": 7.79, "percentage": 77.87, "elapsed_time": "3:25:11", "remaining_time": "0:58:19"}
96
+ {"current_steps": 480, "total_steps": 610, "loss": 0.338, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.397004835582242e-06, "epoch": 7.87, "percentage": 78.69, "elapsed_time": "3:27:21", "remaining_time": "0:56:09"}
97
+ {"current_steps": 485, "total_steps": 610, "loss": 0.3344, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.004018881507016e-06, "epoch": 7.95, "percentage": 79.51, "elapsed_time": "3:29:30", "remaining_time": "0:53:59"}
98
+ {"current_steps": 490, "total_steps": 610, "loss": 0.2856, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.624291562079719e-06, "epoch": 8.03, "percentage": 80.33, "elapsed_time": "3:31:37", "remaining_time": "0:51:49"}
99
+ {"current_steps": 495, "total_steps": 610, "loss": 0.2364, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.258074661184527e-06, "epoch": 8.11, "percentage": 81.15, "elapsed_time": "3:33:46", "remaining_time": "0:49:39"}
100
+ {"current_steps": 500, "total_steps": 610, "loss": 0.2325, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.90561100442036e-06, "epoch": 8.2, "percentage": 81.97, "elapsed_time": "3:35:55", "remaining_time": "0:47:30"}
101
+ {"current_steps": 505, "total_steps": 610, "loss": 0.2289, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.56713429809169e-06, "epoch": 8.28, "percentage": 82.79, "elapsed_time": "3:38:15", "remaining_time": "0:45:22"}
102
+ {"current_steps": 510, "total_steps": 610, "loss": 0.2277, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2428689742461188e-06, "epoch": 8.36, "percentage": 83.61, "elapsed_time": "3:40:24", "remaining_time": "0:43:12"}
103
+ {"current_steps": 515, "total_steps": 610, "loss": 0.2268, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.933030041861312e-06, "epoch": 8.44, "percentage": 84.43, "elapsed_time": "3:42:33", "remaining_time": "0:41:03"}
104
+ {"current_steps": 520, "total_steps": 610, "loss": 0.2264, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.637822944280116e-06, "epoch": 8.52, "percentage": 85.25, "elapsed_time": "3:44:42", "remaining_time": "0:38:53"}
105
+ {"current_steps": 525, "total_steps": 610, "loss": 0.2319, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3574434229882145e-06, "epoch": 8.61, "percentage": 86.07, "elapsed_time": "3:46:51", "remaining_time": "0:36:43"}
106
+ {"current_steps": 530, "total_steps": 610, "loss": 0.2292, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.092077387824884e-06, "epoch": 8.69, "percentage": 86.89, "elapsed_time": "3:49:00", "remaining_time": "0:34:34"}
107
+ {"current_steps": 535, "total_steps": 610, "loss": 0.2261, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8419007937126255e-06, "epoch": 8.77, "percentage": 87.7, "elapsed_time": "3:51:09", "remaining_time": "0:32:24"}
108
+ {"current_steps": 540, "total_steps": 610, "loss": 0.2247, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6070795239876618e-06, "epoch": 8.85, "percentage": 88.52, "elapsed_time": "3:53:18", "remaining_time": "0:30:14"}
109
+ {"current_steps": 545, "total_steps": 610, "loss": 0.2358, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3877692804084685e-06, "epoch": 8.93, "percentage": 89.34, "elapsed_time": "3:55:27", "remaining_time": "0:28:04"}
110
+ {"current_steps": 550, "total_steps": 610, "loss": 0.2207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1841154799154374e-06, "epoch": 9.02, "percentage": 90.16, "elapsed_time": "3:57:33", "remaining_time": "0:25:54"}
111
+ {"current_steps": 555, "total_steps": 610, "loss": 0.1908, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.96253158209956e-07, "epoch": 9.1, "percentage": 90.98, "elapsed_time": "3:59:42", "remaining_time": "0:23:45"}
112
+ {"current_steps": 560, "total_steps": 610, "loss": 0.1906, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.243068802169906e-07, "epoch": 9.18, "percentage": 91.8, "elapsed_time": "4:01:51", "remaining_time": "0:21:35"}
113
+ {"current_steps": 565, "total_steps": 610, "loss": 0.1914, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.683906574904364e-07, "epoch": 9.26, "percentage": 92.62, "elapsed_time": "4:04:00", "remaining_time": "0:19:26"}
114
+ {"current_steps": 570, "total_steps": 610, "loss": 0.1897, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.286078726160549e-07, "epoch": 9.34, "percentage": 93.44, "elapsed_time": "4:06:09", "remaining_time": "0:17:16"}
115
+ {"current_steps": 575, "total_steps": 610, "loss": 0.1926, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.050512106620913e-07, "epoch": 9.43, "percentage": 94.26, "elapsed_time": "4:08:18", "remaining_time": "0:15:06"}
116
+ {"current_steps": 580, "total_steps": 610, "loss": 0.186, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.978025977230736e-07, "epoch": 9.51, "percentage": 95.08, "elapsed_time": "4:10:27", "remaining_time": "0:12:57"}
117
+ {"current_steps": 585, "total_steps": 610, "loss": 0.1863, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0693314659746278e-07, "epoch": 9.59, "percentage": 95.9, "elapsed_time": "4:12:36", "remaining_time": "0:10:47"}
118
+ {"current_steps": 590, "total_steps": 610, "loss": 0.1887, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3250310963527358e-07, "epoch": 9.67, "percentage": 96.72, "elapsed_time": "4:14:45", "remaining_time": "0:08:38"}
119
+ {"current_steps": 595, "total_steps": 610, "loss": 0.1851, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.456183878683243e-08, "epoch": 9.75, "percentage": 97.54, "elapsed_time": "4:16:53", "remaining_time": "0:06:28"}
120
+ {"current_steps": 600, "total_steps": 610, "loss": 0.1906, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.314775287923677e-08, "epoch": 9.84, "percentage": 98.36, "elapsed_time": "4:19:02", "remaining_time": "0:04:19"}
121
+ {"current_steps": 605, "total_steps": 610, "loss": 0.1836, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.288312142173959e-09, "epoch": 9.92, "percentage": 99.18, "elapsed_time": "4:21:21", "remaining_time": "0:02:09"}
122
+ {"current_steps": 610, "total_steps": 610, "loss": 0.1852, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "4:23:28", "remaining_time": "0:00:00"}
123
+ {"current_steps": 610, "total_steps": 610, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "4:23:28", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,884 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 610,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.08,
13
+ "grad_norm": 5.009121417999268,
14
+ "learning_rate": 4.9991711687857826e-05,
15
+ "loss": 3.4699,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.16,
20
+ "grad_norm": 3.122627019882202,
21
+ "learning_rate": 4.9966852247120764e-05,
22
+ "loss": 3.1968,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.25,
27
+ "grad_norm": 2.5967676639556885,
28
+ "learning_rate": 4.9925438161213164e-05,
29
+ "loss": 3.0675,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 0.33,
34
+ "grad_norm": 2.4438161849975586,
35
+ "learning_rate": 4.9867496890364726e-05,
36
+ "loss": 3.0402,
37
+ "step": 20
38
+ },
39
+ {
40
+ "epoch": 0.41,
41
+ "grad_norm": 2.6208345890045166,
42
+ "learning_rate": 4.9793066853402536e-05,
43
+ "loss": 3.0463,
44
+ "step": 25
45
+ },
46
+ {
47
+ "epoch": 0.49,
48
+ "grad_norm": 2.487919330596924,
49
+ "learning_rate": 4.970219740227693e-05,
50
+ "loss": 2.9922,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.57,
55
+ "grad_norm": 2.4597654342651367,
56
+ "learning_rate": 4.9594948789337914e-05,
57
+ "loss": 2.9401,
58
+ "step": 35
59
+ },
60
+ {
61
+ "epoch": 0.66,
62
+ "grad_norm": 2.2669572830200195,
63
+ "learning_rate": 4.947139212738395e-05,
64
+ "loss": 2.913,
65
+ "step": 40
66
+ },
67
+ {
68
+ "epoch": 0.74,
69
+ "grad_norm": 2.3517422676086426,
70
+ "learning_rate": 4.933160934250957e-05,
71
+ "loss": 2.9039,
72
+ "step": 45
73
+ },
74
+ {
75
+ "epoch": 0.82,
76
+ "grad_norm": 2.067209243774414,
77
+ "learning_rate": 4.9175693119783013e-05,
78
+ "loss": 2.9188,
79
+ "step": 50
80
+ },
81
+ {
82
+ "epoch": 0.9,
83
+ "grad_norm": 2.5949714183807373,
84
+ "learning_rate": 4.900374684179004e-05,
85
+ "loss": 2.8791,
86
+ "step": 55
87
+ },
88
+ {
89
+ "epoch": 0.98,
90
+ "grad_norm": 2.3132359981536865,
91
+ "learning_rate": 4.881588452008456e-05,
92
+ "loss": 2.8602,
93
+ "step": 60
94
+ },
95
+ {
96
+ "epoch": 1.07,
97
+ "grad_norm": 2.7244913578033447,
98
+ "learning_rate": 4.861223071959153e-05,
99
+ "loss": 2.6114,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 1.15,
104
+ "grad_norm": 2.426957607269287,
105
+ "learning_rate": 4.839292047601234e-05,
106
+ "loss": 2.5097,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 1.23,
111
+ "grad_norm": 2.518718957901001,
112
+ "learning_rate": 4.815809920628738e-05,
113
+ "loss": 2.497,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 1.31,
118
+ "grad_norm": 2.5183868408203125,
119
+ "learning_rate": 4.790792261217512e-05,
120
+ "loss": 2.4959,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 1.39,
125
+ "grad_norm": 2.5139663219451904,
126
+ "learning_rate": 4.764255657701179e-05,
127
+ "loss": 2.5037,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 1.48,
132
+ "grad_norm": 2.398991823196411,
133
+ "learning_rate": 4.736217705571989e-05,
134
+ "loss": 2.4973,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 1.56,
139
+ "grad_norm": 2.291883707046509,
140
+ "learning_rate": 4.706696995813868e-05,
141
+ "loss": 2.4948,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 1.64,
146
+ "grad_norm": 2.315034866333008,
147
+ "learning_rate": 4.6757131025753886e-05,
148
+ "loss": 2.5013,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 1.72,
153
+ "grad_norm": 2.419813394546509,
154
+ "learning_rate": 4.643286570190831e-05,
155
+ "loss": 2.4919,
156
+ "step": 105
157
+ },
158
+ {
159
+ "epoch": 1.8,
160
+ "grad_norm": 2.408583402633667,
161
+ "learning_rate": 4.609438899557964e-05,
162
+ "loss": 2.4871,
163
+ "step": 110
164
+ },
165
+ {
166
+ "epoch": 1.89,
167
+ "grad_norm": 2.269737958908081,
168
+ "learning_rate": 4.5741925338815474e-05,
169
+ "loss": 2.5035,
170
+ "step": 115
171
+ },
172
+ {
173
+ "epoch": 1.97,
174
+ "grad_norm": 2.33317232131958,
175
+ "learning_rate": 4.5375708437920284e-05,
176
+ "loss": 2.4941,
177
+ "step": 120
178
+ },
179
+ {
180
+ "epoch": 2.05,
181
+ "grad_norm": 2.799999475479126,
182
+ "learning_rate": 4.499598111849299e-05,
183
+ "loss": 2.2503,
184
+ "step": 125
185
+ },
186
+ {
187
+ "epoch": 2.13,
188
+ "grad_norm": 2.7277634143829346,
189
+ "learning_rate": 4.460299516441776e-05,
190
+ "loss": 2.0722,
191
+ "step": 130
192
+ },
193
+ {
194
+ "epoch": 2.21,
195
+ "grad_norm": 2.5570781230926514,
196
+ "learning_rate": 4.4197011150915e-05,
197
+ "loss": 2.0567,
198
+ "step": 135
199
+ },
200
+ {
201
+ "epoch": 2.3,
202
+ "grad_norm": 2.5796124935150146,
203
+ "learning_rate": 4.3778298271762995e-05,
204
+ "loss": 2.0507,
205
+ "step": 140
206
+ },
207
+ {
208
+ "epoch": 2.38,
209
+ "grad_norm": 2.4952049255371094,
210
+ "learning_rate": 4.334713416080498e-05,
211
+ "loss": 2.0694,
212
+ "step": 145
213
+ },
214
+ {
215
+ "epoch": 2.46,
216
+ "grad_norm": 2.6838080883026123,
217
+ "learning_rate": 4.2903804707859835e-05,
218
+ "loss": 2.0711,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 2.54,
223
+ "grad_norm": 3.0429630279541016,
224
+ "learning_rate": 4.2448603869158587e-05,
225
+ "loss": 2.0782,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 2.62,
230
+ "grad_norm": 2.5743329524993896,
231
+ "learning_rate": 4.198183347243233e-05,
232
+ "loss": 2.0982,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 2.7,
237
+ "grad_norm": 2.6105706691741943,
238
+ "learning_rate": 4.1503803016780796e-05,
239
+ "loss": 2.1043,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 2.79,
244
+ "grad_norm": 2.4099245071411133,
245
+ "learning_rate": 4.101482946745439e-05,
246
+ "loss": 2.0925,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 2.87,
251
+ "grad_norm": 2.3642492294311523,
252
+ "learning_rate": 4.051523704568557e-05,
253
+ "loss": 2.1157,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 2.95,
258
+ "grad_norm": 2.626281499862671,
259
+ "learning_rate": 4.000535701370921e-05,
260
+ "loss": 2.1207,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 3.03,
265
+ "grad_norm": 4.121679306030273,
266
+ "learning_rate": 3.948552745511409e-05,
267
+ "loss": 1.9371,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 3.11,
272
+ "grad_norm": 4.904088020324707,
273
+ "learning_rate": 3.895609305067162e-05,
274
+ "loss": 1.6614,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 3.2,
279
+ "grad_norm": 3.118800401687622,
280
+ "learning_rate": 3.841740484979002e-05,
281
+ "loss": 1.6199,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 3.28,
286
+ "grad_norm": 3.1991629600524902,
287
+ "learning_rate": 3.7869820037745776e-05,
288
+ "loss": 1.6182,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 3.36,
293
+ "grad_norm": 2.901879072189331,
294
+ "learning_rate": 3.731370169884662e-05,
295
+ "loss": 1.613,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 3.44,
300
+ "grad_norm": 2.936681032180786,
301
+ "learning_rate": 3.6749418575683e-05,
302
+ "loss": 1.6146,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 3.52,
307
+ "grad_norm": 2.902355194091797,
308
+ "learning_rate": 3.617734482462785e-05,
309
+ "loss": 1.6147,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 3.61,
314
+ "grad_norm": 3.0887398719787598,
315
+ "learning_rate": 3.5597859767746524e-05,
316
+ "loss": 1.6139,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 3.69,
321
+ "grad_norm": 2.7664308547973633,
322
+ "learning_rate": 3.501134764128167e-05,
323
+ "loss": 1.6382,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 3.77,
328
+ "grad_norm": 3.130692720413208,
329
+ "learning_rate": 3.4418197340879635e-05,
330
+ "loss": 1.6468,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 3.85,
335
+ "grad_norm": 2.858790636062622,
336
+ "learning_rate": 3.381880216372738e-05,
337
+ "loss": 1.6518,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 3.93,
342
+ "grad_norm": 2.895685911178589,
343
+ "learning_rate": 3.321355954777087e-05,
344
+ "loss": 1.6574,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 4.02,
349
+ "grad_norm": 4.997915744781494,
350
+ "learning_rate": 3.260287080818795e-05,
351
+ "loss": 1.5766,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 4.1,
356
+ "grad_norm": 5.18801212310791,
357
+ "learning_rate": 3.1987140871290236e-05,
358
+ "loss": 1.2207,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 4.18,
363
+ "grad_norm": 3.488536834716797,
364
+ "learning_rate": 3.136677800603072e-05,
365
+ "loss": 1.202,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 4.26,
370
+ "grad_norm": 3.276165246963501,
371
+ "learning_rate": 3.07421935532949e-05,
372
+ "loss": 1.1862,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 4.34,
377
+ "grad_norm": 3.2313599586486816,
378
+ "learning_rate": 3.0113801653155026e-05,
379
+ "loss": 1.1969,
380
+ "step": 265
381
+ },
382
+ {
383
+ "epoch": 4.43,
384
+ "grad_norm": 3.2551193237304688,
385
+ "learning_rate": 2.9482018970268393e-05,
386
+ "loss": 1.1731,
387
+ "step": 270
388
+ },
389
+ {
390
+ "epoch": 4.51,
391
+ "grad_norm": 3.300238609313965,
392
+ "learning_rate": 2.884726441760155e-05,
393
+ "loss": 1.1747,
394
+ "step": 275
395
+ },
396
+ {
397
+ "epoch": 4.59,
398
+ "grad_norm": 3.164844036102295,
399
+ "learning_rate": 2.8209958878663778e-05,
400
+ "loss": 1.1924,
401
+ "step": 280
402
+ },
403
+ {
404
+ "epoch": 4.67,
405
+ "grad_norm": 3.359332799911499,
406
+ "learning_rate": 2.757052492843401e-05,
407
+ "loss": 1.1844,
408
+ "step": 285
409
+ },
410
+ {
411
+ "epoch": 4.75,
412
+ "grad_norm": 3.1776442527770996,
413
+ "learning_rate": 2.6929386553166164e-05,
414
+ "loss": 1.1892,
415
+ "step": 290
416
+ },
417
+ {
418
+ "epoch": 4.84,
419
+ "grad_norm": 3.2457070350646973,
420
+ "learning_rate": 2.6286968869258665e-05,
421
+ "loss": 1.2005,
422
+ "step": 295
423
+ },
424
+ {
425
+ "epoch": 4.92,
426
+ "grad_norm": 3.3526594638824463,
427
+ "learning_rate": 2.564369784137472e-05,
428
+ "loss": 1.2167,
429
+ "step": 300
430
+ },
431
+ {
432
+ "epoch": 5.0,
433
+ "grad_norm": 3.673542022705078,
434
+ "learning_rate": 2.5e-05,
435
+ "loss": 1.2292,
436
+ "step": 305
437
+ },
438
+ {
439
+ "epoch": 5.08,
440
+ "grad_norm": 4.4025373458862305,
441
+ "learning_rate": 2.4356302158625288e-05,
442
+ "loss": 0.8431,
443
+ "step": 310
444
+ },
445
+ {
446
+ "epoch": 5.16,
447
+ "grad_norm": 3.3656206130981445,
448
+ "learning_rate": 2.3713031130741337e-05,
449
+ "loss": 0.8032,
450
+ "step": 315
451
+ },
452
+ {
453
+ "epoch": 5.25,
454
+ "grad_norm": 3.668022871017456,
455
+ "learning_rate": 2.3070613446833842e-05,
456
+ "loss": 0.8136,
457
+ "step": 320
458
+ },
459
+ {
460
+ "epoch": 5.33,
461
+ "grad_norm": 3.4953699111938477,
462
+ "learning_rate": 2.2429475071565987e-05,
463
+ "loss": 0.813,
464
+ "step": 325
465
+ },
466
+ {
467
+ "epoch": 5.41,
468
+ "grad_norm": 3.4234421253204346,
469
+ "learning_rate": 2.1790041121336225e-05,
470
+ "loss": 0.8105,
471
+ "step": 330
472
+ },
473
+ {
474
+ "epoch": 5.49,
475
+ "grad_norm": 3.4309818744659424,
476
+ "learning_rate": 2.1152735582398452e-05,
477
+ "loss": 0.8095,
478
+ "step": 335
479
+ },
480
+ {
481
+ "epoch": 5.57,
482
+ "grad_norm": 3.3163857460021973,
483
+ "learning_rate": 2.0517981029731616e-05,
484
+ "loss": 0.8083,
485
+ "step": 340
486
+ },
487
+ {
488
+ "epoch": 5.66,
489
+ "grad_norm": 3.3125879764556885,
490
+ "learning_rate": 1.9886198346844987e-05,
491
+ "loss": 0.8329,
492
+ "step": 345
493
+ },
494
+ {
495
+ "epoch": 5.74,
496
+ "grad_norm": 3.3822097778320312,
497
+ "learning_rate": 1.9257806446705116e-05,
498
+ "loss": 0.8162,
499
+ "step": 350
500
+ },
501
+ {
502
+ "epoch": 5.82,
503
+ "grad_norm": 3.432570219039917,
504
+ "learning_rate": 1.8633221993969285e-05,
505
+ "loss": 0.8286,
506
+ "step": 355
507
+ },
508
+ {
509
+ "epoch": 5.9,
510
+ "grad_norm": 3.3860833644866943,
511
+ "learning_rate": 1.8012859128709766e-05,
512
+ "loss": 0.8318,
513
+ "step": 360
514
+ },
515
+ {
516
+ "epoch": 5.98,
517
+ "grad_norm": 3.3469362258911133,
518
+ "learning_rate": 1.7397129191812057e-05,
519
+ "loss": 0.8312,
520
+ "step": 365
521
+ },
522
+ {
523
+ "epoch": 6.07,
524
+ "grad_norm": 3.2420990467071533,
525
+ "learning_rate": 1.6786440452229134e-05,
526
+ "loss": 0.6165,
527
+ "step": 370
528
+ },
529
+ {
530
+ "epoch": 6.15,
531
+ "grad_norm": 3.6129636764526367,
532
+ "learning_rate": 1.618119783627263e-05,
533
+ "loss": 0.5368,
534
+ "step": 375
535
+ },
536
+ {
537
+ "epoch": 6.23,
538
+ "grad_norm": 3.441241979598999,
539
+ "learning_rate": 1.558180265912037e-05,
540
+ "loss": 0.5347,
541
+ "step": 380
542
+ },
543
+ {
544
+ "epoch": 6.31,
545
+ "grad_norm": 3.365767478942871,
546
+ "learning_rate": 1.4988652358718336e-05,
547
+ "loss": 0.5234,
548
+ "step": 385
549
+ },
550
+ {
551
+ "epoch": 6.39,
552
+ "grad_norm": 3.1380422115325928,
553
+ "learning_rate": 1.4402140232253486e-05,
554
+ "loss": 0.514,
555
+ "step": 390
556
+ },
557
+ {
558
+ "epoch": 6.48,
559
+ "grad_norm": 3.0910489559173584,
560
+ "learning_rate": 1.3822655175372149e-05,
561
+ "loss": 0.522,
562
+ "step": 395
563
+ },
564
+ {
565
+ "epoch": 6.56,
566
+ "grad_norm": 3.047086477279663,
567
+ "learning_rate": 1.325058142431701e-05,
568
+ "loss": 0.5207,
569
+ "step": 400
570
+ },
571
+ {
572
+ "epoch": 6.64,
573
+ "grad_norm": 3.230703115463257,
574
+ "learning_rate": 1.2686298301153393e-05,
575
+ "loss": 0.5399,
576
+ "step": 405
577
+ },
578
+ {
579
+ "epoch": 6.72,
580
+ "grad_norm": 3.278348445892334,
581
+ "learning_rate": 1.213017996225424e-05,
582
+ "loss": 0.5304,
583
+ "step": 410
584
+ },
585
+ {
586
+ "epoch": 6.8,
587
+ "grad_norm": 3.2289199829101562,
588
+ "learning_rate": 1.158259515020999e-05,
589
+ "loss": 0.534,
590
+ "step": 415
591
+ },
592
+ {
593
+ "epoch": 6.89,
594
+ "grad_norm": 3.1904938220977783,
595
+ "learning_rate": 1.1043906949328387e-05,
596
+ "loss": 0.5255,
597
+ "step": 420
598
+ },
599
+ {
600
+ "epoch": 6.97,
601
+ "grad_norm": 3.2226648330688477,
602
+ "learning_rate": 1.051447254488591e-05,
603
+ "loss": 0.5388,
604
+ "step": 425
605
+ },
606
+ {
607
+ "epoch": 7.05,
608
+ "grad_norm": 3.1080589294433594,
609
+ "learning_rate": 9.9946429862908e-06,
610
+ "loss": 0.4218,
611
+ "step": 430
612
+ },
613
+ {
614
+ "epoch": 7.13,
615
+ "grad_norm": 3.863534450531006,
616
+ "learning_rate": 9.48476295431443e-06,
617
+ "loss": 0.3403,
618
+ "step": 435
619
+ },
620
+ {
621
+ "epoch": 7.21,
622
+ "grad_norm": 2.675412654876709,
623
+ "learning_rate": 8.985170532545622e-06,
624
+ "loss": 0.3332,
625
+ "step": 440
626
+ },
627
+ {
628
+ "epoch": 7.3,
629
+ "grad_norm": 2.844601631164551,
630
+ "learning_rate": 8.496196983219203e-06,
631
+ "loss": 0.3341,
632
+ "step": 445
633
+ },
634
+ {
635
+ "epoch": 7.38,
636
+ "grad_norm": 2.8453519344329834,
637
+ "learning_rate": 8.018166527567672e-06,
638
+ "loss": 0.3334,
639
+ "step": 450
640
+ },
641
+ {
642
+ "epoch": 7.46,
643
+ "grad_norm": 2.785527229309082,
644
+ "learning_rate": 7.5513961308414065e-06,
645
+ "loss": 0.3329,
646
+ "step": 455
647
+ },
648
+ {
649
+ "epoch": 7.54,
650
+ "grad_norm": 2.7892823219299316,
651
+ "learning_rate": 7.096195292140173e-06,
652
+ "loss": 0.3407,
653
+ "step": 460
654
+ },
655
+ {
656
+ "epoch": 7.62,
657
+ "grad_norm": 2.7924044132232666,
658
+ "learning_rate": 6.652865839195024e-06,
659
+ "loss": 0.3329,
660
+ "step": 465
661
+ },
662
+ {
663
+ "epoch": 7.7,
664
+ "grad_norm": 2.592303991317749,
665
+ "learning_rate": 6.221701728237009e-06,
666
+ "loss": 0.3314,
667
+ "step": 470
668
+ },
669
+ {
670
+ "epoch": 7.79,
671
+ "grad_norm": 2.7689881324768066,
672
+ "learning_rate": 5.8029888490850005e-06,
673
+ "loss": 0.3388,
674
+ "step": 475
675
+ },
676
+ {
677
+ "epoch": 7.87,
678
+ "grad_norm": 2.657968759536743,
679
+ "learning_rate": 5.397004835582242e-06,
680
+ "loss": 0.338,
681
+ "step": 480
682
+ },
683
+ {
684
+ "epoch": 7.95,
685
+ "grad_norm": 2.7076845169067383,
686
+ "learning_rate": 5.004018881507016e-06,
687
+ "loss": 0.3344,
688
+ "step": 485
689
+ },
690
+ {
691
+ "epoch": 8.03,
692
+ "grad_norm": 2.54681396484375,
693
+ "learning_rate": 4.624291562079719e-06,
694
+ "loss": 0.2856,
695
+ "step": 490
696
+ },
697
+ {
698
+ "epoch": 8.11,
699
+ "grad_norm": 2.4071710109710693,
700
+ "learning_rate": 4.258074661184527e-06,
701
+ "loss": 0.2364,
702
+ "step": 495
703
+ },
704
+ {
705
+ "epoch": 8.2,
706
+ "grad_norm": 2.4430058002471924,
707
+ "learning_rate": 3.90561100442036e-06,
708
+ "loss": 0.2325,
709
+ "step": 500
710
+ },
711
+ {
712
+ "epoch": 8.28,
713
+ "grad_norm": 2.2299325466156006,
714
+ "learning_rate": 3.56713429809169e-06,
715
+ "loss": 0.2289,
716
+ "step": 505
717
+ },
718
+ {
719
+ "epoch": 8.36,
720
+ "grad_norm": 2.1451282501220703,
721
+ "learning_rate": 3.2428689742461188e-06,
722
+ "loss": 0.2277,
723
+ "step": 510
724
+ },
725
+ {
726
+ "epoch": 8.44,
727
+ "grad_norm": 2.182375431060791,
728
+ "learning_rate": 2.933030041861312e-06,
729
+ "loss": 0.2268,
730
+ "step": 515
731
+ },
732
+ {
733
+ "epoch": 8.52,
734
+ "grad_norm": 2.242326259613037,
735
+ "learning_rate": 2.637822944280116e-06,
736
+ "loss": 0.2264,
737
+ "step": 520
738
+ },
739
+ {
740
+ "epoch": 8.61,
741
+ "grad_norm": 2.2420380115509033,
742
+ "learning_rate": 2.3574434229882145e-06,
743
+ "loss": 0.2319,
744
+ "step": 525
745
+ },
746
+ {
747
+ "epoch": 8.69,
748
+ "grad_norm": 2.200953960418701,
749
+ "learning_rate": 2.092077387824884e-06,
750
+ "loss": 0.2292,
751
+ "step": 530
752
+ },
753
+ {
754
+ "epoch": 8.77,
755
+ "grad_norm": 2.225543737411499,
756
+ "learning_rate": 1.8419007937126255e-06,
757
+ "loss": 0.2261,
758
+ "step": 535
759
+ },
760
+ {
761
+ "epoch": 8.85,
762
+ "grad_norm": 2.20648193359375,
763
+ "learning_rate": 1.6070795239876618e-06,
764
+ "loss": 0.2247,
765
+ "step": 540
766
+ },
767
+ {
768
+ "epoch": 8.93,
769
+ "grad_norm": 2.1440372467041016,
770
+ "learning_rate": 1.3877692804084685e-06,
771
+ "loss": 0.2358,
772
+ "step": 545
773
+ },
774
+ {
775
+ "epoch": 9.02,
776
+ "grad_norm": 2.046773672103882,
777
+ "learning_rate": 1.1841154799154374e-06,
778
+ "loss": 0.2207,
779
+ "step": 550
780
+ },
781
+ {
782
+ "epoch": 9.1,
783
+ "grad_norm": 1.9511895179748535,
784
+ "learning_rate": 9.96253158209956e-07,
785
+ "loss": 0.1908,
786
+ "step": 555
787
+ },
788
+ {
789
+ "epoch": 9.18,
790
+ "grad_norm": 1.9654351472854614,
791
+ "learning_rate": 8.243068802169906e-07,
792
+ "loss": 0.1906,
793
+ "step": 560
794
+ },
795
+ {
796
+ "epoch": 9.26,
797
+ "grad_norm": 1.9583854675292969,
798
+ "learning_rate": 6.683906574904364e-07,
799
+ "loss": 0.1914,
800
+ "step": 565
801
+ },
802
+ {
803
+ "epoch": 9.34,
804
+ "grad_norm": 1.9551448822021484,
805
+ "learning_rate": 5.286078726160549e-07,
806
+ "loss": 0.1897,
807
+ "step": 570
808
+ },
809
+ {
810
+ "epoch": 9.43,
811
+ "grad_norm": 1.9811819791793823,
812
+ "learning_rate": 4.050512106620913e-07,
813
+ "loss": 0.1926,
814
+ "step": 575
815
+ },
816
+ {
817
+ "epoch": 9.51,
818
+ "grad_norm": 1.927250623703003,
819
+ "learning_rate": 2.978025977230736e-07,
820
+ "loss": 0.186,
821
+ "step": 580
822
+ },
823
+ {
824
+ "epoch": 9.59,
825
+ "grad_norm": 1.8517950773239136,
826
+ "learning_rate": 2.0693314659746278e-07,
827
+ "loss": 0.1863,
828
+ "step": 585
829
+ },
830
+ {
831
+ "epoch": 9.67,
832
+ "grad_norm": 1.9849419593811035,
833
+ "learning_rate": 1.3250310963527358e-07,
834
+ "loss": 0.1887,
835
+ "step": 590
836
+ },
837
+ {
838
+ "epoch": 9.75,
839
+ "grad_norm": 1.8776540756225586,
840
+ "learning_rate": 7.456183878683243e-08,
841
+ "loss": 0.1851,
842
+ "step": 595
843
+ },
844
+ {
845
+ "epoch": 9.84,
846
+ "grad_norm": 2.024672746658325,
847
+ "learning_rate": 3.314775287923677e-08,
848
+ "loss": 0.1906,
849
+ "step": 600
850
+ },
851
+ {
852
+ "epoch": 9.92,
853
+ "grad_norm": 1.9220367670059204,
854
+ "learning_rate": 8.288312142173959e-09,
855
+ "loss": 0.1836,
856
+ "step": 605
857
+ },
858
+ {
859
+ "epoch": 10.0,
860
+ "grad_norm": 2.1801717281341553,
861
+ "learning_rate": 0.0,
862
+ "loss": 0.1852,
863
+ "step": 610
864
+ },
865
+ {
866
+ "epoch": 10.0,
867
+ "step": 610,
868
+ "total_flos": 3.693046753394688e+16,
869
+ "train_loss": 1.2541689154554585,
870
+ "train_runtime": 15808.2669,
871
+ "train_samples_per_second": 1.233,
872
+ "train_steps_per_second": 0.039
873
+ }
874
+ ],
875
+ "logging_steps": 5,
876
+ "max_steps": 610,
877
+ "num_input_tokens_seen": 0,
878
+ "num_train_epochs": 10,
879
+ "save_steps": 100,
880
+ "total_flos": 3.693046753394688e+16,
881
+ "train_batch_size": 4,
882
+ "trial_name": null,
883
+ "trial_params": null
884
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba1a597e37fb804fa12c4ca401d0abf1e479a19ee253f32c55f8203f779bb41d
3
+ size 4667
vocab.json ADDED
The diff for this file is too large to render. See raw diff