Hugging Face's logo

Lorry0727
/
Qwen2-Math-1.5B-MetaMathQA

Text Generation
Transformers
TensorBoard
Safetensors
qwen2
llama-factory
full
Generated from Trainer
conversational
text-generation-inference
Model card Files
xet
Metrics Community
Lorry0727 commited on
Commit
1b92abb
·
verified ·
1 Parent(s): 386fa80

Upload folder using huggingface_hub

Browse files
Files changed (20) hide show
  1. .gitattributes +1 -0
  2. README.md +61 -0
  3. added_tokens.json +5 -0
  4. all_results.json +8 -0
  5. chat_template.jinja +6 -0
  6. config.json +29 -0
  7. generation_config.json +6 -0
  8. merges.txt +0 -0
  9. model.safetensors +3 -0
  10. runs/Jun11_06-18-59_a800/events.out.tfevents.1749622777.a800 +3 -0
  11. runs/Jun11_08-45-19_a800/events.out.tfevents.1749631561.a800 +3 -0
  12. special_tokens_map.json +20 -0
  13. tokenizer.json +3 -0
  14. tokenizer_config.json +44 -0
  15. train_results.json +8 -0
  16. trainer_log.jsonl +309 -0
  17. trainer_state.json +2199 -0
  18. training_args.bin +3 -0
  19. training_loss.png +0 -0
  20. vocab.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: Qwen/Qwen2-Math-1.5B
5
+ tags:
6
+ - llama-factory
7
+ - full
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: sft_lr_5e-6_bs_128
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # sft_lr_1e-6_bs_128
18
+
19
+ This model is a fine-tuned version of [Qwen/Qwen2-Math-1.5B](https://huggingface.co/Qwen/Qwen2-Math-1.5B) on the [meta-math/MetaMathQA](https://huggingface.co/datasets/meta-math/MetaMathQA) dataset.
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-06
39
+ - train_batch_size: 16
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - distributed_type: multi-GPU
43
+ - num_devices: 4
44
+ - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 128
46
+ - total_eval_batch_size: 32
47
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
+ - lr_scheduler_type: cosine
49
+ - lr_scheduler_warmup_ratio: 0.1
50
+ - num_epochs: 1.0
51
+
52
+ ### Training results
53
+
54
+
55
+
56
+ ### Framework versions
57
+
58
+ - Transformers 4.52.4
59
+ - Pytorch 2.7.1+cu126
60
+ - Datasets 3.6.0
61
+ - Tokenizers 0.21.1
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 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 190452489388032.0,
4
+ "train_loss": 0.15608682801641588,
5
+ "train_runtime": 8265.837,
6
+ "train_samples_per_second": 47.787,
7
+ "train_steps_per_second": 0.373
8
+ }
chat_template.jinja ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
2
+ You are a helpful assistant<|im_end|>
3
+ ' }}{% endif %}{{'<|im_start|>' + message['role'] + '
4
+ ' + message['content'] + '<|im_end|>' + '
5
+ '}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
6
+ ' }}{% endif %}
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 151643,
7
+ "eos_token_id": 151643,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 1536,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 8960,
12
+ "max_position_embeddings": 4096,
13
+ "max_window_layers": 21,
14
+ "model_type": "qwen2",
15
+ "num_attention_heads": 12,
16
+ "num_hidden_layers": 28,
17
+ "num_key_value_heads": 2,
18
+ "rms_norm_eps": 1e-06,
19
+ "rope_scaling": null,
20
+ "rope_theta": 10000,
21
+ "sliding_window": 32768,
22
+ "tie_word_embeddings": true,
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.52.4",
25
+ "use_cache": false,
26
+ "use_mrope": false,
27
+ "use_sliding_window": false,
28
+ "vocab_size": 151936
29
+ }
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": "4.52.4"
6
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:506face551b26065554a58cd3fb13d5505dc7e0783f4f76012f92c80d5369adf
3
+ size 3087467144
runs/Jun11_06-18-59_a800/events.out.tfevents.1749622777.a800 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d1549227a23c96d741054f836a11dc89f9bcc68fc2badd66dc8c74d0ba279f5
3
+ size 70708
runs/Jun11_08-45-19_a800/events.out.tfevents.1749631561.a800 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:718ec431daf645572c3f8ca793b2417e4619cefc7bece2f72511c4093164569d
3
+ size 70708
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": "<|im_end|>",
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.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcfe42da0a4497e8b2b172c1f9f4ec423a46dc12907f4349c55025f670422ba9
3
+ size 11418266
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
+ "clean_up_tokenization_spaces": false,
35
+ "eos_token": "<|im_end|>",
36
+ "errors": "replace",
37
+ "extra_special_tokens": {},
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 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 190452489388032.0,
4
+ "train_loss": 0.15608682801641588,
5
+ "train_runtime": 8265.837,
6
+ "train_samples_per_second": 47.787,
7
+ "train_steps_per_second": 0.373
8
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,309 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 3086, "loss": 0.5576, "lr": 1.4563106796116507e-07, "epoch": 0.0032404406999351912, "percentage": 0.32, "elapsed_time": "0:00:30", "remaining_time": "2:34:09"}
2
+ {"current_steps": 20, "total_steps": 3086, "loss": 0.5525, "lr": 3.074433656957929e-07, "epoch": 0.0064808813998703824, "percentage": 0.65, "elapsed_time": "0:00:58", "remaining_time": "2:29:19"}
3
+ {"current_steps": 30, "total_steps": 3086, "loss": 0.5424, "lr": 4.6925566343042074e-07, "epoch": 0.009721322099805573, "percentage": 0.97, "elapsed_time": "0:01:21", "remaining_time": "2:17:51"}
4
+ {"current_steps": 40, "total_steps": 3086, "loss": 0.5232, "lr": 6.310679611650486e-07, "epoch": 0.012961762799740765, "percentage": 1.3, "elapsed_time": "0:01:45", "remaining_time": "2:13:27"}
5
+ {"current_steps": 50, "total_steps": 3086, "loss": 0.4552, "lr": 7.928802588996765e-07, "epoch": 0.016202203499675955, "percentage": 1.62, "elapsed_time": "0:02:10", "remaining_time": "2:12:15"}
6
+ {"current_steps": 60, "total_steps": 3086, "loss": 0.3873, "lr": 9.546925566343043e-07, "epoch": 0.019442644199611146, "percentage": 1.94, "elapsed_time": "0:02:41", "remaining_time": "2:15:45"}
7
+ {"current_steps": 70, "total_steps": 3086, "loss": 0.3333, "lr": 1.1165048543689322e-06, "epoch": 0.02268308489954634, "percentage": 2.27, "elapsed_time": "0:03:07", "remaining_time": "2:14:44"}
8
+ {"current_steps": 80, "total_steps": 3086, "loss": 0.2982, "lr": 1.27831715210356e-06, "epoch": 0.02592352559948153, "percentage": 2.59, "elapsed_time": "0:03:34", "remaining_time": "2:14:36"}
9
+ {"current_steps": 90, "total_steps": 3086, "loss": 0.2797, "lr": 1.440129449838188e-06, "epoch": 0.02916396629941672, "percentage": 2.92, "elapsed_time": "0:04:00", "remaining_time": "2:13:35"}
10
+ {"current_steps": 100, "total_steps": 3086, "loss": 0.2596, "lr": 1.6019417475728158e-06, "epoch": 0.03240440699935191, "percentage": 3.24, "elapsed_time": "0:04:30", "remaining_time": "2:14:36"}
11
+ {"current_steps": 110, "total_steps": 3086, "loss": 0.2563, "lr": 1.7637540453074433e-06, "epoch": 0.0356448476992871, "percentage": 3.56, "elapsed_time": "0:04:58", "remaining_time": "2:14:31"}
12
+ {"current_steps": 120, "total_steps": 3086, "loss": 0.2492, "lr": 1.925566343042071e-06, "epoch": 0.03888528839922229, "percentage": 3.89, "elapsed_time": "0:05:22", "remaining_time": "2:12:44"}
13
+ {"current_steps": 130, "total_steps": 3086, "loss": 0.2451, "lr": 2.0873786407766993e-06, "epoch": 0.04212572909915749, "percentage": 4.21, "elapsed_time": "0:05:52", "remaining_time": "2:13:39"}
14
+ {"current_steps": 140, "total_steps": 3086, "loss": 0.2408, "lr": 2.249190938511327e-06, "epoch": 0.04536616979909268, "percentage": 4.54, "elapsed_time": "0:06:17", "remaining_time": "2:12:25"}
15
+ {"current_steps": 150, "total_steps": 3086, "loss": 0.236, "lr": 2.411003236245955e-06, "epoch": 0.04860661049902787, "percentage": 4.86, "elapsed_time": "0:06:46", "remaining_time": "2:12:45"}
16
+ {"current_steps": 160, "total_steps": 3086, "loss": 0.2246, "lr": 2.5728155339805826e-06, "epoch": 0.05184705119896306, "percentage": 5.18, "elapsed_time": "0:07:14", "remaining_time": "2:12:26"}
17
+ {"current_steps": 170, "total_steps": 3086, "loss": 0.2213, "lr": 2.7346278317152108e-06, "epoch": 0.05508749189889825, "percentage": 5.51, "elapsed_time": "0:07:42", "remaining_time": "2:12:06"}
18
+ {"current_steps": 180, "total_steps": 3086, "loss": 0.2166, "lr": 2.8964401294498386e-06, "epoch": 0.05832793259883344, "percentage": 5.83, "elapsed_time": "0:08:07", "remaining_time": "2:11:10"}
19
+ {"current_steps": 190, "total_steps": 3086, "loss": 0.2119, "lr": 3.058252427184466e-06, "epoch": 0.06156837329876863, "percentage": 6.16, "elapsed_time": "0:08:33", "remaining_time": "2:10:22"}
20
+ {"current_steps": 200, "total_steps": 3086, "loss": 0.2106, "lr": 3.220064724919094e-06, "epoch": 0.06480881399870382, "percentage": 6.48, "elapsed_time": "0:09:01", "remaining_time": "2:10:14"}
21
+ {"current_steps": 210, "total_steps": 3086, "loss": 0.2112, "lr": 3.381877022653722e-06, "epoch": 0.06804925469863901, "percentage": 6.8, "elapsed_time": "0:09:29", "remaining_time": "2:09:59"}
22
+ {"current_steps": 220, "total_steps": 3086, "loss": 0.2042, "lr": 3.54368932038835e-06, "epoch": 0.0712896953985742, "percentage": 7.13, "elapsed_time": "0:09:57", "remaining_time": "2:09:41"}
23
+ {"current_steps": 230, "total_steps": 3086, "loss": 0.1979, "lr": 3.7055016181229774e-06, "epoch": 0.07453013609850939, "percentage": 7.45, "elapsed_time": "0:10:28", "remaining_time": "2:10:07"}
24
+ {"current_steps": 240, "total_steps": 3086, "loss": 0.1992, "lr": 3.867313915857606e-06, "epoch": 0.07777057679844458, "percentage": 7.78, "elapsed_time": "0:10:54", "remaining_time": "2:09:21"}
25
+ {"current_steps": 250, "total_steps": 3086, "loss": 0.1968, "lr": 4.029126213592233e-06, "epoch": 0.08101101749837979, "percentage": 8.1, "elapsed_time": "0:11:20", "remaining_time": "2:08:42"}
26
+ {"current_steps": 260, "total_steps": 3086, "loss": 0.188, "lr": 4.190938511326861e-06, "epoch": 0.08425145819831498, "percentage": 8.43, "elapsed_time": "0:11:46", "remaining_time": "2:08:00"}
27
+ {"current_steps": 270, "total_steps": 3086, "loss": 0.1869, "lr": 4.352750809061489e-06, "epoch": 0.08749189889825017, "percentage": 8.75, "elapsed_time": "0:12:14", "remaining_time": "2:07:35"}
28
+ {"current_steps": 280, "total_steps": 3086, "loss": 0.1873, "lr": 4.514563106796117e-06, "epoch": 0.09073233959818536, "percentage": 9.07, "elapsed_time": "0:12:42", "remaining_time": "2:07:20"}
29
+ {"current_steps": 290, "total_steps": 3086, "loss": 0.1832, "lr": 4.676375404530745e-06, "epoch": 0.09397278029812055, "percentage": 9.4, "elapsed_time": "0:13:08", "remaining_time": "2:06:43"}
30
+ {"current_steps": 300, "total_steps": 3086, "loss": 0.1792, "lr": 4.838187702265373e-06, "epoch": 0.09721322099805574, "percentage": 9.72, "elapsed_time": "0:13:35", "remaining_time": "2:06:08"}
31
+ {"current_steps": 310, "total_steps": 3086, "loss": 0.1741, "lr": 5e-06, "epoch": 0.10045366169799093, "percentage": 10.05, "elapsed_time": "0:14:00", "remaining_time": "2:05:28"}
32
+ {"current_steps": 320, "total_steps": 3086, "loss": 0.1684, "lr": 4.999840024540205e-06, "epoch": 0.10369410239792612, "percentage": 10.37, "elapsed_time": "0:14:26", "remaining_time": "2:04:45"}
33
+ {"current_steps": 330, "total_steps": 3086, "loss": 0.1685, "lr": 4.999360118634534e-06, "epoch": 0.10693454309786131, "percentage": 10.69, "elapsed_time": "0:14:52", "remaining_time": "2:04:12"}
34
+ {"current_steps": 340, "total_steps": 3086, "loss": 0.1656, "lr": 4.998560343701524e-06, "epoch": 0.1101749837977965, "percentage": 11.02, "elapsed_time": "0:15:19", "remaining_time": "2:03:45"}
35
+ {"current_steps": 350, "total_steps": 3086, "loss": 0.1662, "lr": 4.997440802096665e-06, "epoch": 0.11341542449773169, "percentage": 11.34, "elapsed_time": "0:15:48", "remaining_time": "2:03:31"}
36
+ {"current_steps": 360, "total_steps": 3086, "loss": 0.1571, "lr": 4.996001637099303e-06, "epoch": 0.11665586519766688, "percentage": 11.67, "elapsed_time": "0:16:13", "remaining_time": "2:02:52"}
37
+ {"current_steps": 370, "total_steps": 3086, "loss": 0.1618, "lr": 4.994243032894303e-06, "epoch": 0.11989630589760207, "percentage": 11.99, "elapsed_time": "0:16:40", "remaining_time": "2:02:23"}
38
+ {"current_steps": 380, "total_steps": 3086, "loss": 0.165, "lr": 4.992165214548479e-06, "epoch": 0.12313674659753726, "percentage": 12.31, "elapsed_time": "0:17:07", "remaining_time": "2:01:53"}
39
+ {"current_steps": 390, "total_steps": 3086, "loss": 0.1631, "lr": 4.989768447981786e-06, "epoch": 0.12637718729747247, "percentage": 12.64, "elapsed_time": "0:17:33", "remaining_time": "2:01:24"}
40
+ {"current_steps": 400, "total_steps": 3086, "loss": 0.1631, "lr": 4.987053039933294e-06, "epoch": 0.12961762799740764, "percentage": 12.96, "elapsed_time": "0:17:59", "remaining_time": "2:00:49"}
41
+ {"current_steps": 410, "total_steps": 3086, "loss": 0.1596, "lr": 4.984019337921919e-06, "epoch": 0.13285806869734285, "percentage": 13.29, "elapsed_time": "0:18:26", "remaining_time": "2:00:20"}
42
+ {"current_steps": 420, "total_steps": 3086, "loss": 0.1622, "lr": 4.980667730201964e-06, "epoch": 0.13609850939727802, "percentage": 13.61, "elapsed_time": "0:18:53", "remaining_time": "1:59:56"}
43
+ {"current_steps": 430, "total_steps": 3086, "loss": 0.1626, "lr": 4.976998645713417e-06, "epoch": 0.13933895009721323, "percentage": 13.93, "elapsed_time": "0:19:19", "remaining_time": "1:59:24"}
44
+ {"current_steps": 440, "total_steps": 3086, "loss": 0.1563, "lr": 4.973012554027059e-06, "epoch": 0.1425793907971484, "percentage": 14.26, "elapsed_time": "0:19:46", "remaining_time": "1:58:52"}
45
+ {"current_steps": 450, "total_steps": 3086, "loss": 0.1601, "lr": 4.968709965284373e-06, "epoch": 0.1458198314970836, "percentage": 14.58, "elapsed_time": "0:20:13", "remaining_time": "1:58:28"}
46
+ {"current_steps": 460, "total_steps": 3086, "loss": 0.1571, "lr": 4.964091430132246e-06, "epoch": 0.14906027219701878, "percentage": 14.91, "elapsed_time": "0:20:39", "remaining_time": "1:57:58"}
47
+ {"current_steps": 470, "total_steps": 3086, "loss": 0.1578, "lr": 4.959157539652507e-06, "epoch": 0.152300712896954, "percentage": 15.23, "elapsed_time": "0:21:06", "remaining_time": "1:57:30"}
48
+ {"current_steps": 480, "total_steps": 3086, "loss": 0.1564, "lr": 4.9539089252862745e-06, "epoch": 0.15554115359688916, "percentage": 15.55, "elapsed_time": "0:21:31", "remaining_time": "1:56:54"}
49
+ {"current_steps": 490, "total_steps": 3086, "loss": 0.157, "lr": 4.948346258753147e-06, "epoch": 0.15878159429682437, "percentage": 15.88, "elapsed_time": "0:21:57", "remaining_time": "1:56:17"}
50
+ {"current_steps": 500, "total_steps": 3086, "loss": 0.1555, "lr": 4.942470251965232e-06, "epoch": 0.16202203499675957, "percentage": 16.2, "elapsed_time": "0:22:23", "remaining_time": "1:55:46"}
51
+ {"current_steps": 510, "total_steps": 3086, "loss": 0.1536, "lr": 4.936281656936038e-06, "epoch": 0.16526247569669475, "percentage": 16.53, "elapsed_time": "0:22:47", "remaining_time": "1:55:05"}
52
+ {"current_steps": 520, "total_steps": 3086, "loss": 0.1572, "lr": 4.929781265684237e-06, "epoch": 0.16850291639662995, "percentage": 16.85, "elapsed_time": "0:23:15", "remaining_time": "1:54:43"}
53
+ {"current_steps": 530, "total_steps": 3086, "loss": 0.1578, "lr": 4.92296991013229e-06, "epoch": 0.17174335709656513, "percentage": 17.17, "elapsed_time": "0:23:46", "remaining_time": "1:54:37"}
54
+ {"current_steps": 540, "total_steps": 3086, "loss": 0.155, "lr": 4.915848461999987e-06, "epoch": 0.17498379779650033, "percentage": 17.5, "elapsed_time": "0:24:12", "remaining_time": "1:54:07"}
55
+ {"current_steps": 550, "total_steps": 3086, "loss": 0.1539, "lr": 4.908417832692879e-06, "epoch": 0.1782242384964355, "percentage": 17.82, "elapsed_time": "0:24:37", "remaining_time": "1:53:30"}
56
+ {"current_steps": 560, "total_steps": 3086, "loss": 0.1493, "lr": 4.900678973185638e-06, "epoch": 0.18146467919637072, "percentage": 18.15, "elapsed_time": "0:25:01", "remaining_time": "1:52:50"}
57
+ {"current_steps": 570, "total_steps": 3086, "loss": 0.1571, "lr": 4.892632873900349e-06, "epoch": 0.1847051198963059, "percentage": 18.47, "elapsed_time": "0:25:25", "remaining_time": "1:52:12"}
58
+ {"current_steps": 580, "total_steps": 3086, "loss": 0.1546, "lr": 4.884280564579761e-06, "epoch": 0.1879455605962411, "percentage": 18.79, "elapsed_time": "0:25:53", "remaining_time": "1:51:50"}
59
+ {"current_steps": 590, "total_steps": 3086, "loss": 0.1523, "lr": 4.8756231141554925e-06, "epoch": 0.19118600129617627, "percentage": 19.12, "elapsed_time": "0:26:18", "remaining_time": "1:51:19"}
60
+ {"current_steps": 600, "total_steps": 3086, "loss": 0.1497, "lr": 4.866661630611231e-06, "epoch": 0.19442644199611148, "percentage": 19.44, "elapsed_time": "0:26:46", "remaining_time": "1:50:56"}
61
+ {"current_steps": 610, "total_steps": 3086, "loss": 0.1569, "lr": 4.8573972608409395e-06, "epoch": 0.19766688269604665, "percentage": 19.77, "elapsed_time": "0:27:17", "remaining_time": "1:50:47"}
62
+ {"current_steps": 620, "total_steps": 3086, "loss": 0.1506, "lr": 4.847831190502068e-06, "epoch": 0.20090732339598186, "percentage": 20.09, "elapsed_time": "0:27:44", "remaining_time": "1:50:18"}
63
+ {"current_steps": 630, "total_steps": 3086, "loss": 0.1552, "lr": 4.837964643863818e-06, "epoch": 0.20414776409591703, "percentage": 20.41, "elapsed_time": "0:28:09", "remaining_time": "1:49:46"}
64
+ {"current_steps": 640, "total_steps": 3086, "loss": 0.1528, "lr": 4.827798883650455e-06, "epoch": 0.20738820479585224, "percentage": 20.74, "elapsed_time": "0:28:40", "remaining_time": "1:49:33"}
65
+ {"current_steps": 650, "total_steps": 3086, "loss": 0.1537, "lr": 4.817335210879712e-06, "epoch": 0.21062864549578741, "percentage": 21.06, "elapsed_time": "0:29:07", "remaining_time": "1:49:09"}
66
+ {"current_steps": 660, "total_steps": 3086, "loss": 0.1482, "lr": 4.8065749646962815e-06, "epoch": 0.21386908619572262, "percentage": 21.39, "elapsed_time": "0:29:34", "remaining_time": "1:48:43"}
67
+ {"current_steps": 670, "total_steps": 3086, "loss": 0.1524, "lr": 4.795519522200424e-06, "epoch": 0.21710952689565782, "percentage": 21.71, "elapsed_time": "0:30:01", "remaining_time": "1:48:15"}
68
+ {"current_steps": 680, "total_steps": 3086, "loss": 0.1537, "lr": 4.784170298271739e-06, "epoch": 0.220349967595593, "percentage": 22.03, "elapsed_time": "0:30:26", "remaining_time": "1:47:41"}
69
+ {"current_steps": 690, "total_steps": 3086, "loss": 0.1496, "lr": 4.772528745388079e-06, "epoch": 0.2235904082955282, "percentage": 22.36, "elapsed_time": "0:30:51", "remaining_time": "1:47:09"}
70
+ {"current_steps": 700, "total_steps": 3086, "loss": 0.1476, "lr": 4.760596353439665e-06, "epoch": 0.22683084899546338, "percentage": 22.68, "elapsed_time": "0:31:19", "remaining_time": "1:46:45"}
71
+ {"current_steps": 710, "total_steps": 3086, "loss": 0.1505, "lr": 4.748374649538406e-06, "epoch": 0.23007128969539858, "percentage": 23.01, "elapsed_time": "0:31:45", "remaining_time": "1:46:16"}
72
+ {"current_steps": 720, "total_steps": 3086, "loss": 0.1521, "lr": 4.735865197822465e-06, "epoch": 0.23331173039533376, "percentage": 23.33, "elapsed_time": "0:32:13", "remaining_time": "1:45:52"}
73
+ {"current_steps": 730, "total_steps": 3086, "loss": 0.1477, "lr": 4.7230695992560725e-06, "epoch": 0.23655217109526896, "percentage": 23.66, "elapsed_time": "0:32:38", "remaining_time": "1:45:22"}
74
+ {"current_steps": 740, "total_steps": 3086, "loss": 0.1487, "lr": 4.709989491424641e-06, "epoch": 0.23979261179520414, "percentage": 23.98, "elapsed_time": "0:33:07", "remaining_time": "1:45:01"}
75
+ {"current_steps": 750, "total_steps": 3086, "loss": 0.1511, "lr": 4.69662654832518e-06, "epoch": 0.24303305249513935, "percentage": 24.3, "elapsed_time": "0:33:32", "remaining_time": "1:44:29"}
76
+ {"current_steps": 760, "total_steps": 3086, "loss": 0.1545, "lr": 4.682982480152066e-06, "epoch": 0.24627349319507452, "percentage": 24.63, "elapsed_time": "0:33:59", "remaining_time": "1:44:01"}
77
+ {"current_steps": 770, "total_steps": 3086, "loss": 0.1548, "lr": 4.669059033078159e-06, "epoch": 0.24951393389500973, "percentage": 24.95, "elapsed_time": "0:34:26", "remaining_time": "1:43:35"}
78
+ {"current_steps": 780, "total_steps": 3086, "loss": 0.1467, "lr": 4.65485798903134e-06, "epoch": 0.25275437459494493, "percentage": 25.28, "elapsed_time": "0:34:52", "remaining_time": "1:43:06"}
79
+ {"current_steps": 790, "total_steps": 3086, "loss": 0.1458, "lr": 4.640381165466447e-06, "epoch": 0.2559948152948801, "percentage": 25.6, "elapsed_time": "0:35:16", "remaining_time": "1:42:32"}
80
+ {"current_steps": 800, "total_steps": 3086, "loss": 0.1519, "lr": 4.625630415132688e-06, "epoch": 0.2592352559948153, "percentage": 25.92, "elapsed_time": "0:35:44", "remaining_time": "1:42:07"}
81
+ {"current_steps": 810, "total_steps": 3086, "loss": 0.1476, "lr": 4.610607625836515e-06, "epoch": 0.26247569669475046, "percentage": 26.25, "elapsed_time": "0:36:08", "remaining_time": "1:41:34"}
82
+ {"current_steps": 820, "total_steps": 3086, "loss": 0.1469, "lr": 4.595314720200028e-06, "epoch": 0.2657161373946857, "percentage": 26.57, "elapsed_time": "0:36:35", "remaining_time": "1:41:06"}
83
+ {"current_steps": 830, "total_steps": 3086, "loss": 0.1469, "lr": 4.579753655414916e-06, "epoch": 0.26895657809462087, "percentage": 26.9, "elapsed_time": "0:37:02", "remaining_time": "1:40:41"}
84
+ {"current_steps": 840, "total_steps": 3086, "loss": 0.1507, "lr": 4.563926422991973e-06, "epoch": 0.27219701879455604, "percentage": 27.22, "elapsed_time": "0:37:30", "remaining_time": "1:40:16"}
85
+ {"current_steps": 850, "total_steps": 3086, "loss": 0.1505, "lr": 4.5478350485062275e-06, "epoch": 0.2754374594944913, "percentage": 27.54, "elapsed_time": "0:37:56", "remaining_time": "1:39:47"}
86
+ {"current_steps": 860, "total_steps": 3086, "loss": 0.1465, "lr": 4.531481591337706e-06, "epoch": 0.27867790019442645, "percentage": 27.87, "elapsed_time": "0:38:23", "remaining_time": "1:39:22"}
87
+ {"current_steps": 870, "total_steps": 3086, "loss": 0.1497, "lr": 4.51486814440787e-06, "epoch": 0.28191834089436163, "percentage": 28.19, "elapsed_time": "0:38:48", "remaining_time": "1:38:50"}
88
+ {"current_steps": 880, "total_steps": 3086, "loss": 0.1494, "lr": 4.497996833911771e-06, "epoch": 0.2851587815942968, "percentage": 28.52, "elapsed_time": "0:39:15", "remaining_time": "1:38:25"}
89
+ {"current_steps": 890, "total_steps": 3086, "loss": 0.1543, "lr": 4.48086981904593e-06, "epoch": 0.28839922229423204, "percentage": 28.84, "elapsed_time": "0:39:44", "remaining_time": "1:38:02"}
90
+ {"current_steps": 900, "total_steps": 3086, "loss": 0.1414, "lr": 4.463489291732011e-06, "epoch": 0.2916396629941672, "percentage": 29.16, "elapsed_time": "0:40:11", "remaining_time": "1:37:38"}
91
+ {"current_steps": 910, "total_steps": 3086, "loss": 0.1486, "lr": 4.445857476336292e-06, "epoch": 0.2948801036941024, "percentage": 29.49, "elapsed_time": "0:40:37", "remaining_time": "1:37:07"}
92
+ {"current_steps": 920, "total_steps": 3086, "loss": 0.1465, "lr": 4.427976629384994e-06, "epoch": 0.29812054439403757, "percentage": 29.81, "elapsed_time": "0:41:02", "remaining_time": "1:36:37"}
93
+ {"current_steps": 930, "total_steps": 3086, "loss": 0.1468, "lr": 4.409849039275486e-06, "epoch": 0.3013609850939728, "percentage": 30.14, "elapsed_time": "0:41:28", "remaining_time": "1:36:08"}
94
+ {"current_steps": 940, "total_steps": 3086, "loss": 0.1479, "lr": 4.391477025983418e-06, "epoch": 0.304601425793908, "percentage": 30.46, "elapsed_time": "0:41:54", "remaining_time": "1:35:39"}
95
+ {"current_steps": 950, "total_steps": 3086, "loss": 0.1496, "lr": 4.372862940765811e-06, "epoch": 0.30784186649384315, "percentage": 30.78, "elapsed_time": "0:42:22", "remaining_time": "1:35:17"}
96
+ {"current_steps": 960, "total_steps": 3086, "loss": 0.1482, "lr": 4.354009165860136e-06, "epoch": 0.31108230719377833, "percentage": 31.11, "elapsed_time": "0:42:50", "remaining_time": "1:34:53"}
97
+ {"current_steps": 970, "total_steps": 3086, "loss": 0.1435, "lr": 4.3349181141794414e-06, "epoch": 0.31432274789371356, "percentage": 31.43, "elapsed_time": "0:43:17", "remaining_time": "1:34:26"}
98
+ {"current_steps": 980, "total_steps": 3086, "loss": 0.1456, "lr": 4.315592229003544e-06, "epoch": 0.31756318859364874, "percentage": 31.76, "elapsed_time": "0:43:41", "remaining_time": "1:33:54"}
99
+ {"current_steps": 990, "total_steps": 3086, "loss": 0.145, "lr": 4.2960339836663364e-06, "epoch": 0.3208036292935839, "percentage": 32.08, "elapsed_time": "0:44:06", "remaining_time": "1:33:23"}
100
+ {"current_steps": 1000, "total_steps": 3086, "loss": 0.1466, "lr": 4.276245881239253e-06, "epoch": 0.32404406999351915, "percentage": 32.4, "elapsed_time": "0:44:32", "remaining_time": "1:32:55"}
101
+ {"current_steps": 1010, "total_steps": 3086, "loss": 0.1439, "lr": 4.2562304542109186e-06, "epoch": 0.3272845106934543, "percentage": 32.73, "elapsed_time": "0:45:01", "remaining_time": "1:32:31"}
102
+ {"current_steps": 1020, "total_steps": 3086, "loss": 0.1442, "lr": 4.235990264163048e-06, "epoch": 0.3305249513933895, "percentage": 33.05, "elapsed_time": "0:45:25", "remaining_time": "1:32:01"}
103
+ {"current_steps": 1030, "total_steps": 3086, "loss": 0.1433, "lr": 4.215527901442609e-06, "epoch": 0.3337653920933247, "percentage": 33.38, "elapsed_time": "0:45:51", "remaining_time": "1:31:33"}
104
+ {"current_steps": 1040, "total_steps": 3086, "loss": 0.1474, "lr": 4.194845984830308e-06, "epoch": 0.3370058327932599, "percentage": 33.7, "elapsed_time": "0:46:18", "remaining_time": "1:31:05"}
105
+ {"current_steps": 1050, "total_steps": 3086, "loss": 0.149, "lr": 4.173947161205443e-06, "epoch": 0.3402462734931951, "percentage": 34.02, "elapsed_time": "0:46:44", "remaining_time": "1:30:38"}
106
+ {"current_steps": 1060, "total_steps": 3086, "loss": 0.1448, "lr": 4.152834105207146e-06, "epoch": 0.34348671419313026, "percentage": 34.35, "elapsed_time": "0:47:11", "remaining_time": "1:30:11"}
107
+ {"current_steps": 1070, "total_steps": 3086, "loss": 0.1454, "lr": 4.131509518892093e-06, "epoch": 0.34672715489306544, "percentage": 34.67, "elapsed_time": "0:47:37", "remaining_time": "1:29:44"}
108
+ {"current_steps": 1080, "total_steps": 3086, "loss": 0.1469, "lr": 4.109976131388682e-06, "epoch": 0.34996759559300067, "percentage": 35.0, "elapsed_time": "0:48:03", "remaining_time": "1:29:16"}
109
+ {"current_steps": 1090, "total_steps": 3086, "loss": 0.1453, "lr": 4.088236698547767e-06, "epoch": 0.35320803629293585, "percentage": 35.32, "elapsed_time": "0:48:34", "remaining_time": "1:28:56"}
110
+ {"current_steps": 1100, "total_steps": 3086, "loss": 0.1466, "lr": 4.06629400258996e-06, "epoch": 0.356448476992871, "percentage": 35.64, "elapsed_time": "0:49:03", "remaining_time": "1:28:35"}
111
+ {"current_steps": 1110, "total_steps": 3086, "loss": 0.1449, "lr": 4.0441508517495605e-06, "epoch": 0.3596889176928062, "percentage": 35.97, "elapsed_time": "0:49:30", "remaining_time": "1:28:08"}
112
+ {"current_steps": 1120, "total_steps": 3086, "loss": 0.1466, "lr": 4.021810079915159e-06, "epoch": 0.36292935839274143, "percentage": 36.29, "elapsed_time": "0:49:59", "remaining_time": "1:27:45"}
113
+ {"current_steps": 1130, "total_steps": 3086, "loss": 0.1467, "lr": 3.9992745462669515e-06, "epoch": 0.3661697990926766, "percentage": 36.62, "elapsed_time": "0:50:25", "remaining_time": "1:27:17"}
114
+ {"current_steps": 1140, "total_steps": 3086, "loss": 0.1452, "lr": 3.976547134910826e-06, "epoch": 0.3694102397926118, "percentage": 36.94, "elapsed_time": "0:50:52", "remaining_time": "1:26:50"}
115
+ {"current_steps": 1150, "total_steps": 3086, "loss": 0.1478, "lr": 3.953630754509245e-06, "epoch": 0.37265068049254696, "percentage": 37.27, "elapsed_time": "0:51:18", "remaining_time": "1:26:22"}
116
+ {"current_steps": 1160, "total_steps": 3086, "loss": 0.1466, "lr": 3.930528337909002e-06, "epoch": 0.3758911211924822, "percentage": 37.59, "elapsed_time": "0:51:46", "remaining_time": "1:25:57"}
117
+ {"current_steps": 1170, "total_steps": 3086, "loss": 0.1475, "lr": 3.907242841765874e-06, "epoch": 0.37913156189241737, "percentage": 37.91, "elapsed_time": "0:52:13", "remaining_time": "1:25:31"}
118
+ {"current_steps": 1180, "total_steps": 3086, "loss": 0.1386, "lr": 3.883777246166219e-06, "epoch": 0.38237200259235254, "percentage": 38.24, "elapsed_time": "0:52:39", "remaining_time": "1:25:02"}
119
+ {"current_steps": 1190, "total_steps": 3086, "loss": 0.1412, "lr": 3.860134554245597e-06, "epoch": 0.3856124432922878, "percentage": 38.56, "elapsed_time": "0:53:02", "remaining_time": "1:24:31"}
120
+ {"current_steps": 1200, "total_steps": 3086, "loss": 0.1408, "lr": 3.836317791804415e-06, "epoch": 0.38885288399222295, "percentage": 38.89, "elapsed_time": "0:53:29", "remaining_time": "1:24:03"}
121
+ {"current_steps": 1210, "total_steps": 3086, "loss": 0.1428, "lr": 3.81233000692069e-06, "epoch": 0.39209332469215813, "percentage": 39.21, "elapsed_time": "0:53:56", "remaining_time": "1:23:38"}
122
+ {"current_steps": 1220, "total_steps": 3086, "loss": 0.1439, "lr": 3.7881742695599567e-06, "epoch": 0.3953337653920933, "percentage": 39.53, "elapsed_time": "0:54:25", "remaining_time": "1:23:14"}
123
+ {"current_steps": 1230, "total_steps": 3086, "loss": 0.1445, "lr": 3.763853671182367e-06, "epoch": 0.39857420609202854, "percentage": 39.86, "elapsed_time": "0:54:52", "remaining_time": "1:22:48"}
124
+ {"current_steps": 1240, "total_steps": 3086, "loss": 0.1426, "lr": 3.7393713243470475e-06, "epoch": 0.4018146467919637, "percentage": 40.18, "elapsed_time": "0:55:19", "remaining_time": "1:22:21"}
125
+ {"current_steps": 1250, "total_steps": 3086, "loss": 0.1438, "lr": 3.7147303623137508e-06, "epoch": 0.4050550874918989, "percentage": 40.51, "elapsed_time": "0:55:44", "remaining_time": "1:21:52"}
126
+ {"current_steps": 1260, "total_steps": 3086, "loss": 0.1433, "lr": 3.6899339386418635e-06, "epoch": 0.40829552819183407, "percentage": 40.83, "elapsed_time": "0:56:15", "remaining_time": "1:21:31"}
127
+ {"current_steps": 1270, "total_steps": 3086, "loss": 0.1415, "lr": 3.664985226786807e-06, "epoch": 0.4115359688917693, "percentage": 41.15, "elapsed_time": "0:56:40", "remaining_time": "1:21:02"}
128
+ {"current_steps": 1280, "total_steps": 3086, "loss": 0.1424, "lr": 3.6398874196939015e-06, "epoch": 0.4147764095917045, "percentage": 41.48, "elapsed_time": "0:57:06", "remaining_time": "1:20:35"}
129
+ {"current_steps": 1290, "total_steps": 3086, "loss": 0.1434, "lr": 3.614643729389731e-06, "epoch": 0.41801685029163965, "percentage": 41.8, "elapsed_time": "0:57:33", "remaining_time": "1:20:08"}
130
+ {"current_steps": 1300, "total_steps": 3086, "loss": 0.1447, "lr": 3.5892573865710665e-06, "epoch": 0.42125729099157483, "percentage": 42.13, "elapsed_time": "0:57:59", "remaining_time": "1:19:39"}
131
+ {"current_steps": 1310, "total_steps": 3086, "loss": 0.1432, "lr": 3.5637316401913997e-06, "epoch": 0.42449773169151006, "percentage": 42.45, "elapsed_time": "0:58:24", "remaining_time": "1:19:11"}
132
+ {"current_steps": 1320, "total_steps": 3086, "loss": 0.1404, "lr": 3.5380697570451413e-06, "epoch": 0.42773817239144524, "percentage": 42.77, "elapsed_time": "0:58:49", "remaining_time": "1:18:42"}
133
+ {"current_steps": 1330, "total_steps": 3086, "loss": 0.1456, "lr": 3.5122750213495355e-06, "epoch": 0.4309786130913804, "percentage": 43.1, "elapsed_time": "0:59:15", "remaining_time": "1:18:14"}
134
+ {"current_steps": 1340, "total_steps": 3086, "loss": 0.1435, "lr": 3.486350734324346e-06, "epoch": 0.43421905379131565, "percentage": 43.42, "elapsed_time": "0:59:42", "remaining_time": "1:17:48"}
135
+ {"current_steps": 1350, "total_steps": 3086, "loss": 0.1427, "lr": 3.460300213769361e-06, "epoch": 0.4374594944912508, "percentage": 43.75, "elapsed_time": "1:00:11", "remaining_time": "1:17:24"}
136
+ {"current_steps": 1360, "total_steps": 3086, "loss": 0.1403, "lr": 3.4341267936397843e-06, "epoch": 0.440699935191186, "percentage": 44.07, "elapsed_time": "1:00:40", "remaining_time": "1:17:00"}
137
+ {"current_steps": 1370, "total_steps": 3086, "loss": 0.1419, "lr": 3.4078338236195506e-06, "epoch": 0.4439403758911212, "percentage": 44.39, "elapsed_time": "1:01:05", "remaining_time": "1:16:31"}
138
+ {"current_steps": 1380, "total_steps": 3086, "loss": 0.1422, "lr": 3.3814246686926367e-06, "epoch": 0.4471808165910564, "percentage": 44.72, "elapsed_time": "1:01:35", "remaining_time": "1:16:08"}
139
+ {"current_steps": 1390, "total_steps": 3086, "loss": 0.1415, "lr": 3.354902708712402e-06, "epoch": 0.4504212572909916, "percentage": 45.04, "elapsed_time": "1:02:02", "remaining_time": "1:15:41"}
140
+ {"current_steps": 1400, "total_steps": 3086, "loss": 0.1423, "lr": 3.3282713379690418e-06, "epoch": 0.45366169799092676, "percentage": 45.37, "elapsed_time": "1:02:30", "remaining_time": "1:15:16"}
141
+ {"current_steps": 1410, "total_steps": 3086, "loss": 0.144, "lr": 3.301533964755179e-06, "epoch": 0.45690213869086194, "percentage": 45.69, "elapsed_time": "1:02:56", "remaining_time": "1:14:48"}
142
+ {"current_steps": 1420, "total_steps": 3086, "loss": 0.143, "lr": 3.2746940109296744e-06, "epoch": 0.46014257939079717, "percentage": 46.01, "elapsed_time": "1:03:22", "remaining_time": "1:14:20"}
143
+ {"current_steps": 1430, "total_steps": 3086, "loss": 0.1433, "lr": 3.2477549114796892e-06, "epoch": 0.46338302009073234, "percentage": 46.34, "elapsed_time": "1:03:49", "remaining_time": "1:13:54"}
144
+ {"current_steps": 1440, "total_steps": 3086, "loss": 0.1429, "lr": 3.220720114081081e-06, "epoch": 0.4666234607906675, "percentage": 46.66, "elapsed_time": "1:04:16", "remaining_time": "1:13:28"}
145
+ {"current_steps": 1450, "total_steps": 3086, "loss": 0.1447, "lr": 3.193593078657165e-06, "epoch": 0.4698639014906027, "percentage": 46.99, "elapsed_time": "1:04:45", "remaining_time": "1:13:03"}
146
+ {"current_steps": 1460, "total_steps": 3086, "loss": 0.1417, "lr": 3.1663772769359134e-06, "epoch": 0.47310434219053793, "percentage": 47.31, "elapsed_time": "1:05:10", "remaining_time": "1:12:35"}
147
+ {"current_steps": 1470, "total_steps": 3086, "loss": 0.1449, "lr": 3.1390761920056423e-06, "epoch": 0.4763447828904731, "percentage": 47.63, "elapsed_time": "1:05:36", "remaining_time": "1:12:07"}
148
+ {"current_steps": 1480, "total_steps": 3086, "loss": 0.1415, "lr": 3.111693317869241e-06, "epoch": 0.4795852235904083, "percentage": 47.96, "elapsed_time": "1:06:02", "remaining_time": "1:11:40"}
149
+ {"current_steps": 1490, "total_steps": 3086, "loss": 0.1416, "lr": 3.084232158997017e-06, "epoch": 0.48282566429034346, "percentage": 48.28, "elapsed_time": "1:06:27", "remaining_time": "1:11:11"}
150
+ {"current_steps": 1500, "total_steps": 3086, "loss": 0.1366, "lr": 3.0566962298781822e-06, "epoch": 0.4860661049902787, "percentage": 48.61, "elapsed_time": "1:06:57", "remaining_time": "1:10:47"}
151
+ {"current_steps": 1510, "total_steps": 3086, "loss": 0.1427, "lr": 3.0290890545710732e-06, "epoch": 0.48930654569021387, "percentage": 48.93, "elapsed_time": "1:07:24", "remaining_time": "1:10:20"}
152
+ {"current_steps": 1520, "total_steps": 3086, "loss": 0.142, "lr": 3.0014141662521423e-06, "epoch": 0.49254698639014904, "percentage": 49.25, "elapsed_time": "1:07:51", "remaining_time": "1:09:54"}
153
+ {"current_steps": 1530, "total_steps": 3086, "loss": 0.1482, "lr": 2.973675106763776e-06, "epoch": 0.4957874270900843, "percentage": 49.58, "elapsed_time": "1:08:19", "remaining_time": "1:09:29"}
154
+ {"current_steps": 1540, "total_steps": 3086, "loss": 0.1382, "lr": 2.9458754261610113e-06, "epoch": 0.49902786779001945, "percentage": 49.9, "elapsed_time": "1:08:48", "remaining_time": "1:09:04"}
155
+ {"current_steps": 1550, "total_steps": 3086, "loss": 0.1397, "lr": 2.9180186822571953e-06, "epoch": 0.5022683084899546, "percentage": 50.23, "elapsed_time": "1:09:15", "remaining_time": "1:08:37"}
156
+ {"current_steps": 1560, "total_steps": 3086, "loss": 0.1399, "lr": 2.8901084401686635e-06, "epoch": 0.5055087491898899, "percentage": 50.55, "elapsed_time": "1:09:44", "remaining_time": "1:08:13"}
157
+ {"current_steps": 1570, "total_steps": 3086, "loss": 0.1434, "lr": 2.8621482718584615e-06, "epoch": 0.508749189889825, "percentage": 50.87, "elapsed_time": "1:10:10", "remaining_time": "1:07:45"}
158
+ {"current_steps": 1580, "total_steps": 3086, "loss": 0.1391, "lr": 2.8341417556792163e-06, "epoch": 0.5119896305897602, "percentage": 51.2, "elapsed_time": "1:10:37", "remaining_time": "1:07:18"}
159
+ {"current_steps": 1590, "total_steps": 3086, "loss": 0.1369, "lr": 2.8060924759151687e-06, "epoch": 0.5152300712896954, "percentage": 51.52, "elapsed_time": "1:11:02", "remaining_time": "1:06:50"}
160
+ {"current_steps": 1600, "total_steps": 3086, "loss": 0.1428, "lr": 2.7780040223234623e-06, "epoch": 0.5184705119896306, "percentage": 51.85, "elapsed_time": "1:11:28", "remaining_time": "1:06:23"}
161
+ {"current_steps": 1610, "total_steps": 3086, "loss": 0.1422, "lr": 2.749879989674718e-06, "epoch": 0.5217109526895658, "percentage": 52.17, "elapsed_time": "1:11:56", "remaining_time": "1:05:57"}
162
+ {"current_steps": 1620, "total_steps": 3086, "loss": 0.1356, "lr": 2.72172397729298e-06, "epoch": 0.5249513933895009, "percentage": 52.5, "elapsed_time": "1:12:24", "remaining_time": "1:05:31"}
163
+ {"current_steps": 1630, "total_steps": 3086, "loss": 0.1421, "lr": 2.6935395885950705e-06, "epoch": 0.5281918340894362, "percentage": 52.82, "elapsed_time": "1:12:50", "remaining_time": "1:05:04"}
164
+ {"current_steps": 1640, "total_steps": 3086, "loss": 0.1397, "lr": 2.665330430629421e-06, "epoch": 0.5314322747893714, "percentage": 53.14, "elapsed_time": "1:13:16", "remaining_time": "1:04:36"}
165
+ {"current_steps": 1650, "total_steps": 3086, "loss": 0.1388, "lr": 2.6371001136144458e-06, "epoch": 0.5346727154893065, "percentage": 53.47, "elapsed_time": "1:13:42", "remaining_time": "1:04:09"}
166
+ {"current_steps": 1660, "total_steps": 3086, "loss": 0.1412, "lr": 2.6088522504764983e-06, "epoch": 0.5379131561892417, "percentage": 53.79, "elapsed_time": "1:14:07", "remaining_time": "1:03:40"}
167
+ {"current_steps": 1670, "total_steps": 3086, "loss": 0.1392, "lr": 2.580590456387497e-06, "epoch": 0.541153596889177, "percentage": 54.12, "elapsed_time": "1:14:33", "remaining_time": "1:03:12"}
168
+ {"current_steps": 1680, "total_steps": 3086, "loss": 0.142, "lr": 2.5523183483022406e-06, "epoch": 0.5443940375891121, "percentage": 54.44, "elapsed_time": "1:15:00", "remaining_time": "1:02:46"}
169
+ {"current_steps": 1690, "total_steps": 3086, "loss": 0.14, "lr": 2.5240395444955255e-06, "epoch": 0.5476344782890473, "percentage": 54.76, "elapsed_time": "1:15:25", "remaining_time": "1:02:18"}
170
+ {"current_steps": 1700, "total_steps": 3086, "loss": 0.1422, "lr": 2.495757664099063e-06, "epoch": 0.5508749189889826, "percentage": 55.09, "elapsed_time": "1:15:53", "remaining_time": "1:01:52"}
171
+ {"current_steps": 1710, "total_steps": 3086, "loss": 0.1399, "lr": 2.4674763266383096e-06, "epoch": 0.5541153596889177, "percentage": 55.41, "elapsed_time": "1:16:20", "remaining_time": "1:01:26"}
172
+ {"current_steps": 1720, "total_steps": 3086, "loss": 0.1389, "lr": 2.439199151569236e-06, "epoch": 0.5573558003888529, "percentage": 55.74, "elapsed_time": "1:16:47", "remaining_time": "1:00:58"}
173
+ {"current_steps": 1730, "total_steps": 3086, "loss": 0.1364, "lr": 2.41092975781511e-06, "epoch": 0.560596241088788, "percentage": 56.06, "elapsed_time": "1:17:13", "remaining_time": "1:00:31"}
174
+ {"current_steps": 1740, "total_steps": 3086, "loss": 0.1356, "lr": 2.382671763303342e-06, "epoch": 0.5638366817887233, "percentage": 56.38, "elapsed_time": "1:17:40", "remaining_time": "1:00:05"}
175
+ {"current_steps": 1750, "total_steps": 3086, "loss": 0.1399, "lr": 2.3544287845024643e-06, "epoch": 0.5670771224886585, "percentage": 56.71, "elapsed_time": "1:18:06", "remaining_time": "0:59:37"}
176
+ {"current_steps": 1760, "total_steps": 3086, "loss": 0.1442, "lr": 2.326204435959291e-06, "epoch": 0.5703175631885936, "percentage": 57.03, "elapsed_time": "1:18:32", "remaining_time": "0:59:10"}
177
+ {"current_steps": 1770, "total_steps": 3086, "loss": 0.142, "lr": 2.2980023298363334e-06, "epoch": 0.5735580038885288, "percentage": 57.36, "elapsed_time": "1:19:01", "remaining_time": "0:58:45"}
178
+ {"current_steps": 1780, "total_steps": 3086, "loss": 0.1386, "lr": 2.2698260754495057e-06, "epoch": 0.5767984445884641, "percentage": 57.68, "elapsed_time": "1:19:30", "remaining_time": "0:58:19"}
179
+ {"current_steps": 1790, "total_steps": 3086, "loss": 0.1379, "lr": 2.2416792788062073e-06, "epoch": 0.5800388852883992, "percentage": 58.0, "elapsed_time": "1:19:57", "remaining_time": "0:57:53"}
180
+ {"current_steps": 1800, "total_steps": 3086, "loss": 0.1394, "lr": 2.213565542143828e-06, "epoch": 0.5832793259883344, "percentage": 58.33, "elapsed_time": "1:20:25", "remaining_time": "0:57:27"}
181
+ {"current_steps": 1810, "total_steps": 3086, "loss": 0.1361, "lr": 2.1854884634687267e-06, "epoch": 0.5865197666882696, "percentage": 58.65, "elapsed_time": "1:20:52", "remaining_time": "0:57:00"}
182
+ {"current_steps": 1820, "total_steps": 3086, "loss": 0.134, "lr": 2.1574516360957588e-06, "epoch": 0.5897602073882048, "percentage": 58.98, "elapsed_time": "1:21:18", "remaining_time": "0:56:33"}
183
+ {"current_steps": 1830, "total_steps": 3086, "loss": 0.1378, "lr": 2.129458648188405e-06, "epoch": 0.59300064808814, "percentage": 59.3, "elapsed_time": "1:21:44", "remaining_time": "0:56:06"}
184
+ {"current_steps": 1840, "total_steps": 3086, "loss": 0.1387, "lr": 2.1015130822995553e-06, "epoch": 0.5962410887880751, "percentage": 59.62, "elapsed_time": "1:22:11", "remaining_time": "0:55:39"}
185
+ {"current_steps": 1850, "total_steps": 3086, "loss": 0.1353, "lr": 2.0736185149130117e-06, "epoch": 0.5994815294880104, "percentage": 59.95, "elapsed_time": "1:22:38", "remaining_time": "0:55:12"}
186
+ {"current_steps": 1860, "total_steps": 3086, "loss": 0.142, "lr": 2.0457785159857673e-06, "epoch": 0.6027219701879456, "percentage": 60.27, "elapsed_time": "1:23:05", "remaining_time": "0:54:46"}
187
+ {"current_steps": 1870, "total_steps": 3086, "loss": 0.1371, "lr": 2.0179966484911268e-06, "epoch": 0.6059624108878807, "percentage": 60.6, "elapsed_time": "1:23:32", "remaining_time": "0:54:19"}
188
+ {"current_steps": 1880, "total_steps": 3086, "loss": 0.1377, "lr": 1.990276467962711e-06, "epoch": 0.609202851587816, "percentage": 60.92, "elapsed_time": "1:23:58", "remaining_time": "0:53:51"}
189
+ {"current_steps": 1890, "total_steps": 3086, "loss": 0.1403, "lr": 1.9626215220394213e-06, "epoch": 0.6124432922877512, "percentage": 61.24, "elapsed_time": "1:24:24", "remaining_time": "0:53:24"}
190
+ {"current_steps": 1900, "total_steps": 3086, "loss": 0.1369, "lr": 1.9350353500114072e-06, "epoch": 0.6156837329876863, "percentage": 61.57, "elapsed_time": "1:24:49", "remaining_time": "0:52:57"}
191
+ {"current_steps": 1910, "total_steps": 3086, "loss": 0.1406, "lr": 1.907521482367114e-06, "epoch": 0.6189241736876215, "percentage": 61.89, "elapsed_time": "1:25:15", "remaining_time": "0:52:29"}
192
+ {"current_steps": 1920, "total_steps": 3086, "loss": 0.137, "lr": 1.8800834403414437e-06, "epoch": 0.6221646143875567, "percentage": 62.22, "elapsed_time": "1:25:41", "remaining_time": "0:52:02"}
193
+ {"current_steps": 1930, "total_steps": 3086, "loss": 0.1362, "lr": 1.8527247354651048e-06, "epoch": 0.6254050550874919, "percentage": 62.54, "elapsed_time": "1:26:05", "remaining_time": "0:51:34"}
194
+ {"current_steps": 1940, "total_steps": 3086, "loss": 0.1378, "lr": 1.8254488691152145e-06, "epoch": 0.6286454957874271, "percentage": 62.86, "elapsed_time": "1:26:30", "remaining_time": "0:51:05"}
195
+ {"current_steps": 1950, "total_steps": 3086, "loss": 0.1387, "lr": 1.7982593320671793e-06, "epoch": 0.6318859364873622, "percentage": 63.19, "elapsed_time": "1:26:55", "remaining_time": "0:50:38"}
196
+ {"current_steps": 1960, "total_steps": 3086, "loss": 0.1378, "lr": 1.7711596040479523e-06, "epoch": 0.6351263771872975, "percentage": 63.51, "elapsed_time": "1:27:22", "remaining_time": "0:50:11"}
197
+ {"current_steps": 1970, "total_steps": 3086, "loss": 0.1382, "lr": 1.744153153290693e-06, "epoch": 0.6383668178872327, "percentage": 63.84, "elapsed_time": "1:27:49", "remaining_time": "0:49:45"}
198
+ {"current_steps": 1980, "total_steps": 3086, "loss": 0.1398, "lr": 1.717243436090904e-06, "epoch": 0.6416072585871678, "percentage": 64.16, "elapsed_time": "1:28:17", "remaining_time": "0:49:18"}
199
+ {"current_steps": 1990, "total_steps": 3086, "loss": 0.1342, "lr": 1.6904338963640913e-06, "epoch": 0.6448476992871031, "percentage": 64.48, "elapsed_time": "1:28:46", "remaining_time": "0:48:53"}
200
+ {"current_steps": 2000, "total_steps": 3086, "loss": 0.1384, "lr": 1.6637279652050103e-06, "epoch": 0.6480881399870383, "percentage": 64.81, "elapsed_time": "1:29:15", "remaining_time": "0:48:27"}
201
+ {"current_steps": 2010, "total_steps": 3086, "loss": 0.1345, "lr": 1.6371290604485532e-06, "epoch": 0.6513285806869734, "percentage": 65.13, "elapsed_time": "1:29:41", "remaining_time": "0:48:00"}
202
+ {"current_steps": 2020, "total_steps": 3086, "loss": 0.136, "lr": 1.610640586232336e-06, "epoch": 0.6545690213869086, "percentage": 65.46, "elapsed_time": "1:30:06", "remaining_time": "0:47:33"}
203
+ {"current_steps": 2030, "total_steps": 3086, "loss": 0.1339, "lr": 1.5842659325610326e-06, "epoch": 0.6578094620868438, "percentage": 65.78, "elapsed_time": "1:30:33", "remaining_time": "0:47:06"}
204
+ {"current_steps": 2040, "total_steps": 3086, "loss": 0.132, "lr": 1.5580084748725193e-06, "epoch": 0.661049902786779, "percentage": 66.1, "elapsed_time": "1:31:00", "remaining_time": "0:46:39"}
205
+ {"current_steps": 2050, "total_steps": 3086, "loss": 0.1419, "lr": 1.5318715736058926e-06, "epoch": 0.6642903434867142, "percentage": 66.43, "elapsed_time": "1:31:29", "remaining_time": "0:46:14"}
206
+ {"current_steps": 2060, "total_steps": 3086, "loss": 0.1377, "lr": 1.5058585737713883e-06, "epoch": 0.6675307841866494, "percentage": 66.75, "elapsed_time": "1:31:55", "remaining_time": "0:45:46"}
207
+ {"current_steps": 2070, "total_steps": 3086, "loss": 0.1373, "lr": 1.4799728045222956e-06, "epoch": 0.6707712248865846, "percentage": 67.08, "elapsed_time": "1:32:20", "remaining_time": "0:45:19"}
208
+ {"current_steps": 2080, "total_steps": 3086, "loss": 0.1396, "lr": 1.4542175787288818e-06, "epoch": 0.6740116655865198, "percentage": 67.4, "elapsed_time": "1:32:44", "remaining_time": "0:44:51"}
209
+ {"current_steps": 2090, "total_steps": 3086, "loss": 0.139, "lr": 1.428596192554421e-06, "epoch": 0.6772521062864549, "percentage": 67.73, "elapsed_time": "1:33:12", "remaining_time": "0:44:25"}
210
+ {"current_steps": 2100, "total_steps": 3086, "loss": 0.1386, "lr": 1.4031119250333393e-06, "epoch": 0.6804925469863902, "percentage": 68.05, "elapsed_time": "1:33:38", "remaining_time": "0:43:58"}
211
+ {"current_steps": 2110, "total_steps": 3086, "loss": 0.138, "lr": 1.3777680376515667e-06, "epoch": 0.6837329876863253, "percentage": 68.37, "elapsed_time": "1:34:04", "remaining_time": "0:43:30"}
212
+ {"current_steps": 2120, "total_steps": 3086, "loss": 0.1374, "lr": 1.3525677739291338e-06, "epoch": 0.6869734283862605, "percentage": 68.7, "elapsed_time": "1:34:31", "remaining_time": "0:43:04"}
213
+ {"current_steps": 2130, "total_steps": 3086, "loss": 0.1391, "lr": 1.3275143590050622e-06, "epoch": 0.6902138690861958, "percentage": 69.02, "elapsed_time": "1:34:56", "remaining_time": "0:42:36"}
214
+ {"current_steps": 2140, "total_steps": 3086, "loss": 0.1356, "lr": 1.3026109992246084e-06, "epoch": 0.6934543097861309, "percentage": 69.35, "elapsed_time": "1:35:24", "remaining_time": "0:42:10"}
215
+ {"current_steps": 2150, "total_steps": 3086, "loss": 0.1365, "lr": 1.2778608817289167e-06, "epoch": 0.6966947504860661, "percentage": 69.67, "elapsed_time": "1:35:53", "remaining_time": "0:41:44"}
216
+ {"current_steps": 2160, "total_steps": 3086, "loss": 0.1411, "lr": 1.2532671740471292e-06, "epoch": 0.6999351911860013, "percentage": 69.99, "elapsed_time": "1:36:19", "remaining_time": "0:41:17"}
217
+ {"current_steps": 2170, "total_steps": 3086, "loss": 0.1396, "lr": 1.2288330236910025e-06, "epoch": 0.7031756318859365, "percentage": 70.32, "elapsed_time": "1:36:48", "remaining_time": "0:40:51"}
218
+ {"current_steps": 2180, "total_steps": 3086, "loss": 0.1394, "lr": 1.204561557752085e-06, "epoch": 0.7064160725858717, "percentage": 70.64, "elapsed_time": "1:37:16", "remaining_time": "0:40:25"}
219
+ {"current_steps": 2190, "total_steps": 3086, "loss": 0.1365, "lr": 1.1804558825015172e-06, "epoch": 0.7096565132858069, "percentage": 70.97, "elapsed_time": "1:37:43", "remaining_time": "0:39:58"}
220
+ {"current_steps": 2200, "total_steps": 3086, "loss": 0.1362, "lr": 1.1565190829924838e-06, "epoch": 0.712896953985742, "percentage": 71.29, "elapsed_time": "1:38:09", "remaining_time": "0:39:31"}
221
+ {"current_steps": 2210, "total_steps": 3086, "loss": 0.1398, "lr": 1.13275422266539e-06, "epoch": 0.7161373946856773, "percentage": 71.61, "elapsed_time": "1:38:37", "remaining_time": "0:39:05"}
222
+ {"current_steps": 2220, "total_steps": 3086, "loss": 0.1383, "lr": 1.109164342955803e-06, "epoch": 0.7193778353856124, "percentage": 71.94, "elapsed_time": "1:39:04", "remaining_time": "0:38:38"}
223
+ {"current_steps": 2230, "total_steps": 3086, "loss": 0.1373, "lr": 1.0857524629052057e-06, "epoch": 0.7226182760855476, "percentage": 72.26, "elapsed_time": "1:39:31", "remaining_time": "0:38:12"}
224
+ {"current_steps": 2240, "total_steps": 3086, "loss": 0.1376, "lr": 1.0625215787746185e-06, "epoch": 0.7258587167854829, "percentage": 72.59, "elapsed_time": "1:39:57", "remaining_time": "0:37:45"}
225
+ {"current_steps": 2250, "total_steps": 3086, "loss": 0.1327, "lr": 1.0394746636611364e-06, "epoch": 0.729099157485418, "percentage": 72.91, "elapsed_time": "1:40:25", "remaining_time": "0:37:18"}
226
+ {"current_steps": 2260, "total_steps": 3086, "loss": 0.1362, "lr": 1.016614667117434e-06, "epoch": 0.7323395981853532, "percentage": 73.23, "elapsed_time": "1:40:53", "remaining_time": "0:36:52"}
227
+ {"current_steps": 2270, "total_steps": 3086, "loss": 0.1408, "lr": 9.939445147742786e-07, "epoch": 0.7355800388852884, "percentage": 73.56, "elapsed_time": "1:41:19", "remaining_time": "0:36:25"}
228
+ {"current_steps": 2280, "total_steps": 3086, "loss": 0.1383, "lr": 9.714671079661048e-07, "epoch": 0.7388204795852236, "percentage": 73.88, "elapsed_time": "1:41:42", "remaining_time": "0:35:57"}
229
+ {"current_steps": 2290, "total_steps": 3086, "loss": 0.1367, "lr": 9.491853233597037e-07, "epoch": 0.7420609202851588, "percentage": 74.21, "elapsed_time": "1:42:08", "remaining_time": "0:35:30"}
230
+ {"current_steps": 2300, "total_steps": 3086, "loss": 0.1367, "lr": 9.271020125860658e-07, "epoch": 0.7453013609850939, "percentage": 74.53, "elapsed_time": "1:42:34", "remaining_time": "0:35:03"}
231
+ {"current_steps": 2310, "total_steps": 3086, "loss": 0.1364, "lr": 9.052200018754276e-07, "epoch": 0.7485418016850292, "percentage": 74.85, "elapsed_time": "1:42:58", "remaining_time": "0:34:35"}
232
+ {"current_steps": 2320, "total_steps": 3086, "loss": 0.1376, "lr": 8.835420916955686e-07, "epoch": 0.7517822423849644, "percentage": 75.18, "elapsed_time": "1:43:23", "remaining_time": "0:34:08"}
233
+ {"current_steps": 2330, "total_steps": 3086, "loss": 0.1374, "lr": 8.620710563934063e-07, "epoch": 0.7550226830848995, "percentage": 75.5, "elapsed_time": "1:43:49", "remaining_time": "0:33:41"}
234
+ {"current_steps": 2340, "total_steps": 3086, "loss": 0.1369, "lr": 8.408096438399371e-07, "epoch": 0.7582631237848347, "percentage": 75.83, "elapsed_time": "1:44:17", "remaining_time": "0:33:14"}
235
+ {"current_steps": 2350, "total_steps": 3086, "loss": 0.1353, "lr": 8.19760575078562e-07, "epoch": 0.76150356448477, "percentage": 76.15, "elapsed_time": "1:44:45", "remaining_time": "0:32:48"}
236
+ {"current_steps": 2360, "total_steps": 3086, "loss": 0.1336, "lr": 7.989265439768412e-07, "epoch": 0.7647440051847051, "percentage": 76.47, "elapsed_time": "1:45:14", "remaining_time": "0:32:22"}
237
+ {"current_steps": 2370, "total_steps": 3086, "loss": 0.1338, "lr": 7.783102168817406e-07, "epoch": 0.7679844458846403, "percentage": 76.8, "elapsed_time": "1:45:42", "remaining_time": "0:31:56"}
238
+ {"current_steps": 2380, "total_steps": 3086, "loss": 0.1371, "lr": 7.579142322783837e-07, "epoch": 0.7712248865845756, "percentage": 77.12, "elapsed_time": "1:46:11", "remaining_time": "0:31:30"}
239
+ {"current_steps": 2390, "total_steps": 3086, "loss": 0.1333, "lr": 7.37741200452384e-07, "epoch": 0.7744653272845107, "percentage": 77.45, "elapsed_time": "1:46:36", "remaining_time": "0:31:02"}
240
+ {"current_steps": 2400, "total_steps": 3086, "loss": 0.1384, "lr": 7.177937031557733e-07, "epoch": 0.7777057679844459, "percentage": 77.77, "elapsed_time": "1:47:03", "remaining_time": "0:30:36"}
241
+ {"current_steps": 2410, "total_steps": 3086, "loss": 0.1304, "lr": 6.98074293276595e-07, "epoch": 0.780946208684381, "percentage": 78.09, "elapsed_time": "1:47:29", "remaining_time": "0:30:08"}
242
+ {"current_steps": 2420, "total_steps": 3086, "loss": 0.1371, "lr": 6.785854945121773e-07, "epoch": 0.7841866493843163, "percentage": 78.42, "elapsed_time": "1:47:55", "remaining_time": "0:29:42"}
243
+ {"current_steps": 2430, "total_steps": 3086, "loss": 0.1357, "lr": 6.593298010461541e-07, "epoch": 0.7874270900842515, "percentage": 78.74, "elapsed_time": "1:48:20", "remaining_time": "0:29:14"}
244
+ {"current_steps": 2440, "total_steps": 3086, "loss": 0.1388, "lr": 6.403096772292611e-07, "epoch": 0.7906675307841866, "percentage": 79.07, "elapsed_time": "1:48:46", "remaining_time": "0:28:47"}
245
+ {"current_steps": 2450, "total_steps": 3086, "loss": 0.1383, "lr": 6.215275572639387e-07, "epoch": 0.7939079714841218, "percentage": 79.39, "elapsed_time": "1:49:13", "remaining_time": "0:28:21"}
246
+ {"current_steps": 2460, "total_steps": 3086, "loss": 0.1374, "lr": 6.029858448928091e-07, "epoch": 0.7971484121840571, "percentage": 79.71, "elapsed_time": "1:49:42", "remaining_time": "0:27:55"}
247
+ {"current_steps": 2470, "total_steps": 3086, "loss": 0.1394, "lr": 5.846869130910413e-07, "epoch": 0.8003888528839922, "percentage": 80.04, "elapsed_time": "1:50:08", "remaining_time": "0:27:28"}
248
+ {"current_steps": 2480, "total_steps": 3086, "loss": 0.1345, "lr": 5.666331037626585e-07, "epoch": 0.8036292935839274, "percentage": 80.36, "elapsed_time": "1:50:34", "remaining_time": "0:27:01"}
249
+ {"current_steps": 2490, "total_steps": 3086, "loss": 0.1349, "lr": 5.488267274408207e-07, "epoch": 0.8068697342838627, "percentage": 80.69, "elapsed_time": "1:51:02", "remaining_time": "0:26:34"}
250
+ {"current_steps": 2500, "total_steps": 3086, "loss": 0.1325, "lr": 5.312700629921181e-07, "epoch": 0.8101101749837978, "percentage": 81.01, "elapsed_time": "1:51:29", "remaining_time": "0:26:07"}
251
+ {"current_steps": 2510, "total_steps": 3086, "loss": 0.1333, "lr": 5.139653573249253e-07, "epoch": 0.813350615683733, "percentage": 81.34, "elapsed_time": "1:51:59", "remaining_time": "0:25:41"}
252
+ {"current_steps": 2520, "total_steps": 3086, "loss": 0.1342, "lr": 4.969148251018385e-07, "epoch": 0.8165910563836681, "percentage": 81.66, "elapsed_time": "1:52:23", "remaining_time": "0:25:14"}
253
+ {"current_steps": 2530, "total_steps": 3086, "loss": 0.1352, "lr": 4.801206484562446e-07, "epoch": 0.8198314970836034, "percentage": 81.98, "elapsed_time": "1:52:51", "remaining_time": "0:24:48"}
254
+ {"current_steps": 2540, "total_steps": 3086, "loss": 0.1357, "lr": 4.6358497671304623e-07, "epoch": 0.8230719377835386, "percentage": 82.31, "elapsed_time": "1:53:18", "remaining_time": "0:24:21"}
255
+ {"current_steps": 2550, "total_steps": 3086, "loss": 0.1388, "lr": 4.473099261135977e-07, "epoch": 0.8263123784834737, "percentage": 82.63, "elapsed_time": "1:53:44", "remaining_time": "0:23:54"}
256
+ {"current_steps": 2560, "total_steps": 3086, "loss": 0.1364, "lr": 4.312975795448593e-07, "epoch": 0.829552819183409, "percentage": 82.96, "elapsed_time": "1:54:10", "remaining_time": "0:23:27"}
257
+ {"current_steps": 2570, "total_steps": 3086, "loss": 0.1315, "lr": 4.1554998627283655e-07, "epoch": 0.8327932598833442, "percentage": 83.28, "elapsed_time": "1:54:36", "remaining_time": "0:23:00"}
258
+ {"current_steps": 2580, "total_steps": 3086, "loss": 0.1413, "lr": 4.000691616803076e-07, "epoch": 0.8360337005832793, "percentage": 83.6, "elapsed_time": "1:55:01", "remaining_time": "0:22:33"}
259
+ {"current_steps": 2590, "total_steps": 3086, "loss": 0.1364, "lr": 3.8485708700889913e-07, "epoch": 0.8392741412832145, "percentage": 83.93, "elapsed_time": "1:55:29", "remaining_time": "0:22:06"}
260
+ {"current_steps": 2600, "total_steps": 3086, "loss": 0.1346, "lr": 3.699157091055225e-07, "epoch": 0.8425145819831497, "percentage": 84.25, "elapsed_time": "1:55:56", "remaining_time": "0:21:40"}
261
+ {"current_steps": 2610, "total_steps": 3086, "loss": 0.1379, "lr": 3.552469401732181e-07, "epoch": 0.8457550226830849, "percentage": 84.58, "elapsed_time": "1:56:22", "remaining_time": "0:21:13"}
262
+ {"current_steps": 2620, "total_steps": 3086, "loss": 0.1407, "lr": 3.408526575264298e-07, "epoch": 0.8489954633830201, "percentage": 84.9, "elapsed_time": "1:56:49", "remaining_time": "0:20:46"}
263
+ {"current_steps": 2630, "total_steps": 3086, "loss": 0.1341, "lr": 3.267347033507456e-07, "epoch": 0.8522359040829552, "percentage": 85.22, "elapsed_time": "1:57:16", "remaining_time": "0:20:19"}
264
+ {"current_steps": 2640, "total_steps": 3086, "loss": 0.1354, "lr": 3.128948844671334e-07, "epoch": 0.8554763447828905, "percentage": 85.55, "elapsed_time": "1:57:41", "remaining_time": "0:19:52"}
265
+ {"current_steps": 2650, "total_steps": 3086, "loss": 0.136, "lr": 2.993349721007041e-07, "epoch": 0.8587167854828257, "percentage": 85.87, "elapsed_time": "1:58:06", "remaining_time": "0:19:25"}
266
+ {"current_steps": 2660, "total_steps": 3086, "loss": 0.1379, "lr": 2.860567016540322e-07, "epoch": 0.8619572261827608, "percentage": 86.2, "elapsed_time": "1:58:34", "remaining_time": "0:18:59"}
267
+ {"current_steps": 2670, "total_steps": 3086, "loss": 0.138, "lr": 2.730617724850523e-07, "epoch": 0.8651976668826961, "percentage": 86.52, "elapsed_time": "1:59:00", "remaining_time": "0:18:32"}
268
+ {"current_steps": 2680, "total_steps": 3086, "loss": 0.1372, "lr": 2.603518476895794e-07, "epoch": 0.8684381075826313, "percentage": 86.84, "elapsed_time": "1:59:26", "remaining_time": "0:18:05"}
269
+ {"current_steps": 2690, "total_steps": 3086, "loss": 0.1315, "lr": 2.479285538884649e-07, "epoch": 0.8716785482825664, "percentage": 87.17, "elapsed_time": "1:59:52", "remaining_time": "0:17:38"}
270
+ {"current_steps": 2700, "total_steps": 3086, "loss": 0.1395, "lr": 2.3579348101941817e-07, "epoch": 0.8749189889825016, "percentage": 87.49, "elapsed_time": "2:00:18", "remaining_time": "0:17:11"}
271
+ {"current_steps": 2710, "total_steps": 3086, "loss": 0.1346, "lr": 2.239481821335296e-07, "epoch": 0.8781594296824368, "percentage": 87.82, "elapsed_time": "2:00:43", "remaining_time": "0:16:44"}
272
+ {"current_steps": 2720, "total_steps": 3086, "loss": 0.135, "lr": 2.1239417319650707e-07, "epoch": 0.881399870382372, "percentage": 88.14, "elapsed_time": "2:01:07", "remaining_time": "0:16:17"}
273
+ {"current_steps": 2730, "total_steps": 3086, "loss": 0.1386, "lr": 2.0113293289466484e-07, "epoch": 0.8846403110823072, "percentage": 88.46, "elapsed_time": "2:01:33", "remaining_time": "0:15:51"}
274
+ {"current_steps": 2740, "total_steps": 3086, "loss": 0.1374, "lr": 1.901659024456784e-07, "epoch": 0.8878807517822424, "percentage": 88.79, "elapsed_time": "2:02:00", "remaining_time": "0:15:24"}
275
+ {"current_steps": 2750, "total_steps": 3086, "loss": 0.1349, "lr": 1.7949448541413923e-07, "epoch": 0.8911211924821776, "percentage": 89.11, "elapsed_time": "2:02:25", "remaining_time": "0:14:57"}
276
+ {"current_steps": 2760, "total_steps": 3086, "loss": 0.1372, "lr": 1.6912004753192463e-07, "epoch": 0.8943616331821128, "percentage": 89.44, "elapsed_time": "2:02:52", "remaining_time": "0:14:30"}
277
+ {"current_steps": 2770, "total_steps": 3086, "loss": 0.1398, "lr": 1.5904391652341034e-07, "epoch": 0.8976020738820479, "percentage": 89.76, "elapsed_time": "2:03:17", "remaining_time": "0:14:03"}
278
+ {"current_steps": 2780, "total_steps": 3086, "loss": 0.137, "lr": 1.492673819355489e-07, "epoch": 0.9008425145819832, "percentage": 90.08, "elapsed_time": "2:03:43", "remaining_time": "0:13:37"}
279
+ {"current_steps": 2790, "total_steps": 3086, "loss": 0.1344, "lr": 1.3979169497283369e-07, "epoch": 0.9040829552819183, "percentage": 90.41, "elapsed_time": "2:04:08", "remaining_time": "0:13:10"}
280
+ {"current_steps": 2800, "total_steps": 3086, "loss": 0.1351, "lr": 1.3061806833716773e-07, "epoch": 0.9073233959818535, "percentage": 90.73, "elapsed_time": "2:04:34", "remaining_time": "0:12:43"}
281
+ {"current_steps": 2810, "total_steps": 3086, "loss": 0.1381, "lr": 1.2174767607266214e-07, "epoch": 0.9105638366817888, "percentage": 91.06, "elapsed_time": "2:04:59", "remaining_time": "0:12:16"}
282
+ {"current_steps": 2820, "total_steps": 3086, "loss": 0.1409, "lr": 1.1318165341538096e-07, "epoch": 0.9138042773817239, "percentage": 91.38, "elapsed_time": "2:05:25", "remaining_time": "0:11:49"}
283
+ {"current_steps": 2830, "total_steps": 3086, "loss": 0.1357, "lr": 1.0492109664805572e-07, "epoch": 0.9170447180816591, "percentage": 91.7, "elapsed_time": "2:05:50", "remaining_time": "0:11:23"}
284
+ {"current_steps": 2840, "total_steps": 3086, "loss": 0.1367, "lr": 9.69670629597802e-08, "epoch": 0.9202851587815943, "percentage": 92.03, "elapsed_time": "2:06:17", "remaining_time": "0:10:56"}
285
+ {"current_steps": 2850, "total_steps": 3086, "loss": 0.1367, "lr": 8.932057031071113e-08, "epoch": 0.9235255994815295, "percentage": 92.35, "elapsed_time": "2:06:44", "remaining_time": "0:10:29"}
286
+ {"current_steps": 2860, "total_steps": 3086, "loss": 0.1372, "lr": 8.198259730178987e-08, "epoch": 0.9267660401814647, "percentage": 92.68, "elapsed_time": "2:07:11", "remaining_time": "0:10:03"}
287
+ {"current_steps": 2870, "total_steps": 3086, "loss": 0.1362, "lr": 7.495408304950213e-08, "epoch": 0.9300064808813999, "percentage": 93.0, "elapsed_time": "2:07:38", "remaining_time": "0:09:36"}
288
+ {"current_steps": 2880, "total_steps": 3086, "loss": 0.1364, "lr": 6.823592706568738e-08, "epoch": 0.933246921581335, "percentage": 93.32, "elapsed_time": "2:08:04", "remaining_time": "0:09:09"}
289
+ {"current_steps": 2890, "total_steps": 3086, "loss": 0.1397, "lr": 6.182898914241897e-08, "epoch": 0.9364873622812703, "percentage": 93.65, "elapsed_time": "2:08:33", "remaining_time": "0:08:43"}
290
+ {"current_steps": 2900, "total_steps": 3086, "loss": 0.1355, "lr": 5.5734089241969434e-08, "epoch": 0.9397278029812054, "percentage": 93.97, "elapsed_time": "2:09:01", "remaining_time": "0:08:16"}
291
+ {"current_steps": 2910, "total_steps": 3086, "loss": 0.1353, "lr": 4.995200739186967e-08, "epoch": 0.9429682436811406, "percentage": 94.3, "elapsed_time": "2:09:28", "remaining_time": "0:07:49"}
292
+ {"current_steps": 2920, "total_steps": 3086, "loss": 0.1349, "lr": 4.448348358508187e-08, "epoch": 0.9462086843810759, "percentage": 94.62, "elapsed_time": "2:09:53", "remaining_time": "0:07:23"}
293
+ {"current_steps": 2930, "total_steps": 3086, "loss": 0.1391, "lr": 3.932921768529413e-08, "epoch": 0.949449125081011, "percentage": 94.94, "elapsed_time": "2:10:20", "remaining_time": "0:06:56"}
294
+ {"current_steps": 2940, "total_steps": 3086, "loss": 0.1408, "lr": 3.448986933735299e-08, "epoch": 0.9526895657809462, "percentage": 95.27, "elapsed_time": "2:10:46", "remaining_time": "0:06:29"}
295
+ {"current_steps": 2950, "total_steps": 3086, "loss": 0.1376, "lr": 2.996605788283924e-08, "epoch": 0.9559300064808814, "percentage": 95.59, "elapsed_time": "2:11:12", "remaining_time": "0:06:02"}
296
+ {"current_steps": 2960, "total_steps": 3086, "loss": 0.1323, "lr": 2.5758362280807192e-08, "epoch": 0.9591704471808166, "percentage": 95.92, "elapsed_time": "2:11:39", "remaining_time": "0:05:36"}
297
+ {"current_steps": 2970, "total_steps": 3086, "loss": 0.1349, "lr": 2.1867321033687827e-08, "epoch": 0.9624108878807518, "percentage": 96.24, "elapsed_time": "2:12:04", "remaining_time": "0:05:09"}
298
+ {"current_steps": 2980, "total_steps": 3086, "loss": 0.1358, "lr": 1.8293432118371146e-08, "epoch": 0.9656513285806869, "percentage": 96.57, "elapsed_time": "2:12:31", "remaining_time": "0:04:42"}
299
+ {"current_steps": 2990, "total_steps": 3086, "loss": 0.1406, "lr": 1.5037152922475207e-08, "epoch": 0.9688917692806222, "percentage": 96.89, "elapsed_time": "2:12:58", "remaining_time": "0:04:16"}
300
+ {"current_steps": 3000, "total_steps": 3086, "loss": 0.1362, "lr": 1.2098900185808781e-08, "epoch": 0.9721322099805574, "percentage": 97.21, "elapsed_time": "2:13:25", "remaining_time": "0:03:49"}
301
+ {"current_steps": 3010, "total_steps": 3086, "loss": 0.1355, "lr": 9.479049947038743e-09, "epoch": 0.9753726506804925, "percentage": 97.54, "elapsed_time": "2:13:53", "remaining_time": "0:03:22"}
302
+ {"current_steps": 3020, "total_steps": 3086, "loss": 0.134, "lr": 7.177937495561616e-09, "epoch": 0.9786130913804277, "percentage": 97.86, "elapsed_time": "2:14:18", "remaining_time": "0:02:56"}
303
+ {"current_steps": 3030, "total_steps": 3086, "loss": 0.1367, "lr": 5.1958573285959546e-09, "epoch": 0.981853532080363, "percentage": 98.19, "elapsed_time": "2:14:43", "remaining_time": "0:02:29"}
304
+ {"current_steps": 3040, "total_steps": 3086, "loss": 0.1353, "lr": 3.533063113490276e-09, "epoch": 0.9850939727802981, "percentage": 98.51, "elapsed_time": "2:15:10", "remaining_time": "0:02:02"}
305
+ {"current_steps": 3050, "total_steps": 3086, "loss": 0.132, "lr": 2.189767655259578e-09, "epoch": 0.9883344134802333, "percentage": 98.83, "elapsed_time": "2:15:35", "remaining_time": "0:01:36"}
306
+ {"current_steps": 3060, "total_steps": 3086, "loss": 0.1306, "lr": 1.1661428693512944e-09, "epoch": 0.9915748541801686, "percentage": 99.16, "elapsed_time": "2:16:00", "remaining_time": "0:01:09"}
307
+ {"current_steps": 3070, "total_steps": 3086, "loss": 0.1314, "lr": 4.623197596417828e-10, "epoch": 0.9948152948801037, "percentage": 99.48, "elapsed_time": "2:16:25", "remaining_time": "0:00:42"}
308
+ {"current_steps": 3080, "total_steps": 3086, "loss": 0.1368, "lr": 7.838840167168071e-11, "epoch": 0.9980557355800389, "percentage": 99.81, "elapsed_time": "2:16:54", "remaining_time": "0:00:16"}
309
+ {"current_steps": 3086, "total_steps": 3086, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "2:17:45", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,2199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3086,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0032404406999351912,
14
+ "grad_norm": 13.30365312552572,
15
+ "learning_rate": 1.4563106796116507e-07,
16
+ "loss": 0.5576,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0064808813998703824,
21
+ "grad_norm": 14.204922777466475,
22
+ "learning_rate": 3.074433656957929e-07,
23
+ "loss": 0.5525,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.009721322099805573,
28
+ "grad_norm": 9.138075119736614,
29
+ "learning_rate": 4.6925566343042074e-07,
30
+ "loss": 0.5424,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.012961762799740765,
35
+ "grad_norm": 5.051726436122705,
36
+ "learning_rate": 6.310679611650486e-07,
37
+ "loss": 0.5232,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.016202203499675955,
42
+ "grad_norm": 3.46444651483391,
43
+ "learning_rate": 7.928802588996765e-07,
44
+ "loss": 0.4552,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.019442644199611146,
49
+ "grad_norm": 1.6233777031065288,
50
+ "learning_rate": 9.546925566343043e-07,
51
+ "loss": 0.3873,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.02268308489954634,
56
+ "grad_norm": 1.1762000179825056,
57
+ "learning_rate": 1.1165048543689322e-06,
58
+ "loss": 0.3333,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.02592352559948153,
63
+ "grad_norm": 0.9558551467742782,
64
+ "learning_rate": 1.27831715210356e-06,
65
+ "loss": 0.2982,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.02916396629941672,
70
+ "grad_norm": 0.9422801874700582,
71
+ "learning_rate": 1.440129449838188e-06,
72
+ "loss": 0.2797,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.03240440699935191,
77
+ "grad_norm": 0.8214754997776478,
78
+ "learning_rate": 1.6019417475728158e-06,
79
+ "loss": 0.2596,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.0356448476992871,
84
+ "grad_norm": 0.9532933362931321,
85
+ "learning_rate": 1.7637540453074433e-06,
86
+ "loss": 0.2563,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.03888528839922229,
91
+ "grad_norm": 0.9390014348115954,
92
+ "learning_rate": 1.925566343042071e-06,
93
+ "loss": 0.2492,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.04212572909915749,
98
+ "grad_norm": 0.8940098493565622,
99
+ "learning_rate": 2.0873786407766993e-06,
100
+ "loss": 0.2451,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.04536616979909268,
105
+ "grad_norm": 0.956678618390739,
106
+ "learning_rate": 2.249190938511327e-06,
107
+ "loss": 0.2408,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.04860661049902787,
112
+ "grad_norm": 0.9610225933792697,
113
+ "learning_rate": 2.411003236245955e-06,
114
+ "loss": 0.236,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.05184705119896306,
119
+ "grad_norm": 0.9269340800650678,
120
+ "learning_rate": 2.5728155339805826e-06,
121
+ "loss": 0.2246,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.05508749189889825,
126
+ "grad_norm": 0.9180868113485474,
127
+ "learning_rate": 2.7346278317152108e-06,
128
+ "loss": 0.2213,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.05832793259883344,
133
+ "grad_norm": 0.9786641616216852,
134
+ "learning_rate": 2.8964401294498386e-06,
135
+ "loss": 0.2166,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.06156837329876863,
140
+ "grad_norm": 0.980551217904519,
141
+ "learning_rate": 3.058252427184466e-06,
142
+ "loss": 0.2119,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.06480881399870382,
147
+ "grad_norm": 0.8626959550478579,
148
+ "learning_rate": 3.220064724919094e-06,
149
+ "loss": 0.2106,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.06804925469863901,
154
+ "grad_norm": 0.8489552811589715,
155
+ "learning_rate": 3.381877022653722e-06,
156
+ "loss": 0.2112,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.0712896953985742,
161
+ "grad_norm": 0.8916159222304701,
162
+ "learning_rate": 3.54368932038835e-06,
163
+ "loss": 0.2042,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.07453013609850939,
168
+ "grad_norm": 0.9086960370554139,
169
+ "learning_rate": 3.7055016181229774e-06,
170
+ "loss": 0.1979,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.07777057679844458,
175
+ "grad_norm": 0.8413464808652309,
176
+ "learning_rate": 3.867313915857606e-06,
177
+ "loss": 0.1992,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.08101101749837979,
182
+ "grad_norm": 0.8989628261518818,
183
+ "learning_rate": 4.029126213592233e-06,
184
+ "loss": 0.1968,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.08425145819831498,
189
+ "grad_norm": 0.9955980487760288,
190
+ "learning_rate": 4.190938511326861e-06,
191
+ "loss": 0.188,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.08749189889825017,
196
+ "grad_norm": 0.9206294362335926,
197
+ "learning_rate": 4.352750809061489e-06,
198
+ "loss": 0.1869,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.09073233959818536,
203
+ "grad_norm": 0.8143016067643809,
204
+ "learning_rate": 4.514563106796117e-06,
205
+ "loss": 0.1873,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.09397278029812055,
210
+ "grad_norm": 0.840197832466951,
211
+ "learning_rate": 4.676375404530745e-06,
212
+ "loss": 0.1832,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.09721322099805574,
217
+ "grad_norm": 0.8031811731017143,
218
+ "learning_rate": 4.838187702265373e-06,
219
+ "loss": 0.1792,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.10045366169799093,
224
+ "grad_norm": 0.7566604882873896,
225
+ "learning_rate": 5e-06,
226
+ "loss": 0.1741,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.10369410239792612,
231
+ "grad_norm": 0.6800771883014759,
232
+ "learning_rate": 4.999840024540205e-06,
233
+ "loss": 0.1684,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.10693454309786131,
238
+ "grad_norm": 0.5814963126209718,
239
+ "learning_rate": 4.999360118634534e-06,
240
+ "loss": 0.1685,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.1101749837977965,
245
+ "grad_norm": 0.5139919383970122,
246
+ "learning_rate": 4.998560343701524e-06,
247
+ "loss": 0.1656,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.11341542449773169,
252
+ "grad_norm": 0.4719697466406009,
253
+ "learning_rate": 4.997440802096665e-06,
254
+ "loss": 0.1662,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.11665586519766688,
259
+ "grad_norm": 0.5441236604152835,
260
+ "learning_rate": 4.996001637099303e-06,
261
+ "loss": 0.1571,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.11989630589760207,
266
+ "grad_norm": 0.5924545330957118,
267
+ "learning_rate": 4.994243032894303e-06,
268
+ "loss": 0.1618,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.12313674659753726,
273
+ "grad_norm": 0.4776889270665497,
274
+ "learning_rate": 4.992165214548479e-06,
275
+ "loss": 0.165,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.12637718729747247,
280
+ "grad_norm": 0.4865661174720737,
281
+ "learning_rate": 4.989768447981786e-06,
282
+ "loss": 0.1631,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.12961762799740764,
287
+ "grad_norm": 0.39533484163777743,
288
+ "learning_rate": 4.987053039933294e-06,
289
+ "loss": 0.1631,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.13285806869734285,
294
+ "grad_norm": 0.4194333976090903,
295
+ "learning_rate": 4.984019337921919e-06,
296
+ "loss": 0.1596,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.13609850939727802,
301
+ "grad_norm": 0.417606467091398,
302
+ "learning_rate": 4.980667730201964e-06,
303
+ "loss": 0.1622,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.13933895009721323,
308
+ "grad_norm": 0.4135543560101031,
309
+ "learning_rate": 4.976998645713417e-06,
310
+ "loss": 0.1626,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.1425793907971484,
315
+ "grad_norm": 0.3863220511507732,
316
+ "learning_rate": 4.973012554027059e-06,
317
+ "loss": 0.1563,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.1458198314970836,
322
+ "grad_norm": 0.4057624631990764,
323
+ "learning_rate": 4.968709965284373e-06,
324
+ "loss": 0.1601,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.14906027219701878,
329
+ "grad_norm": 0.43679600565915516,
330
+ "learning_rate": 4.964091430132246e-06,
331
+ "loss": 0.1571,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.152300712896954,
336
+ "grad_norm": 0.410392605815529,
337
+ "learning_rate": 4.959157539652507e-06,
338
+ "loss": 0.1578,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.15554115359688916,
343
+ "grad_norm": 0.4269387014050368,
344
+ "learning_rate": 4.9539089252862745e-06,
345
+ "loss": 0.1564,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.15878159429682437,
350
+ "grad_norm": 0.3867586094696986,
351
+ "learning_rate": 4.948346258753147e-06,
352
+ "loss": 0.157,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.16202203499675957,
357
+ "grad_norm": 0.371324732751771,
358
+ "learning_rate": 4.942470251965232e-06,
359
+ "loss": 0.1555,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.16526247569669475,
364
+ "grad_norm": 0.38710422014101686,
365
+ "learning_rate": 4.936281656936038e-06,
366
+ "loss": 0.1536,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.16850291639662995,
371
+ "grad_norm": 0.396987101322912,
372
+ "learning_rate": 4.929781265684237e-06,
373
+ "loss": 0.1572,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.17174335709656513,
378
+ "grad_norm": 0.4019858569286597,
379
+ "learning_rate": 4.92296991013229e-06,
380
+ "loss": 0.1578,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.17498379779650033,
385
+ "grad_norm": 0.38858464136588106,
386
+ "learning_rate": 4.915848461999987e-06,
387
+ "loss": 0.155,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.1782242384964355,
392
+ "grad_norm": 0.4114170865581837,
393
+ "learning_rate": 4.908417832692879e-06,
394
+ "loss": 0.1539,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.18146467919637072,
399
+ "grad_norm": 0.40190769617100247,
400
+ "learning_rate": 4.900678973185638e-06,
401
+ "loss": 0.1493,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.1847051198963059,
406
+ "grad_norm": 0.3776026088033294,
407
+ "learning_rate": 4.892632873900349e-06,
408
+ "loss": 0.1571,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.1879455605962411,
413
+ "grad_norm": 0.3872354376930627,
414
+ "learning_rate": 4.884280564579761e-06,
415
+ "loss": 0.1546,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.19118600129617627,
420
+ "grad_norm": 0.40614075304432395,
421
+ "learning_rate": 4.8756231141554925e-06,
422
+ "loss": 0.1523,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.19442644199611148,
427
+ "grad_norm": 0.3841160702323512,
428
+ "learning_rate": 4.866661630611231e-06,
429
+ "loss": 0.1497,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.19766688269604665,
434
+ "grad_norm": 0.37858054721421236,
435
+ "learning_rate": 4.8573972608409395e-06,
436
+ "loss": 0.1569,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.20090732339598186,
441
+ "grad_norm": 0.4017068446743905,
442
+ "learning_rate": 4.847831190502068e-06,
443
+ "loss": 0.1506,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.20414776409591703,
448
+ "grad_norm": 0.386610704005529,
449
+ "learning_rate": 4.837964643863818e-06,
450
+ "loss": 0.1552,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.20738820479585224,
455
+ "grad_norm": 0.3611796883060248,
456
+ "learning_rate": 4.827798883650455e-06,
457
+ "loss": 0.1528,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.21062864549578741,
462
+ "grad_norm": 0.36233797273951096,
463
+ "learning_rate": 4.817335210879712e-06,
464
+ "loss": 0.1537,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.21386908619572262,
469
+ "grad_norm": 0.355034084310162,
470
+ "learning_rate": 4.8065749646962815e-06,
471
+ "loss": 0.1482,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.21710952689565782,
476
+ "grad_norm": 0.3714493090743044,
477
+ "learning_rate": 4.795519522200424e-06,
478
+ "loss": 0.1524,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.220349967595593,
483
+ "grad_norm": 0.3595872341149548,
484
+ "learning_rate": 4.784170298271739e-06,
485
+ "loss": 0.1537,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.2235904082955282,
490
+ "grad_norm": 0.3657197570676038,
491
+ "learning_rate": 4.772528745388079e-06,
492
+ "loss": 0.1496,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.22683084899546338,
497
+ "grad_norm": 0.39757465881991577,
498
+ "learning_rate": 4.760596353439665e-06,
499
+ "loss": 0.1476,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.23007128969539858,
504
+ "grad_norm": 0.3725684613588135,
505
+ "learning_rate": 4.748374649538406e-06,
506
+ "loss": 0.1505,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.23331173039533376,
511
+ "grad_norm": 0.373958535247375,
512
+ "learning_rate": 4.735865197822465e-06,
513
+ "loss": 0.1521,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.23655217109526896,
518
+ "grad_norm": 0.3895028323791198,
519
+ "learning_rate": 4.7230695992560725e-06,
520
+ "loss": 0.1477,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.23979261179520414,
525
+ "grad_norm": 0.362297190858576,
526
+ "learning_rate": 4.709989491424641e-06,
527
+ "loss": 0.1487,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.24303305249513935,
532
+ "grad_norm": 0.3501236992769411,
533
+ "learning_rate": 4.69662654832518e-06,
534
+ "loss": 0.1511,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.24627349319507452,
539
+ "grad_norm": 0.396141744926481,
540
+ "learning_rate": 4.682982480152066e-06,
541
+ "loss": 0.1545,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.24951393389500973,
546
+ "grad_norm": 0.386092381571303,
547
+ "learning_rate": 4.669059033078159e-06,
548
+ "loss": 0.1548,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.25275437459494493,
553
+ "grad_norm": 0.37841627886408896,
554
+ "learning_rate": 4.65485798903134e-06,
555
+ "loss": 0.1467,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.2559948152948801,
560
+ "grad_norm": 0.3790829553236213,
561
+ "learning_rate": 4.640381165466447e-06,
562
+ "loss": 0.1458,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.2592352559948153,
567
+ "grad_norm": 0.3676425416717676,
568
+ "learning_rate": 4.625630415132688e-06,
569
+ "loss": 0.1519,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.26247569669475046,
574
+ "grad_norm": 0.4020941562750785,
575
+ "learning_rate": 4.610607625836515e-06,
576
+ "loss": 0.1476,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.2657161373946857,
581
+ "grad_norm": 0.36503070299017704,
582
+ "learning_rate": 4.595314720200028e-06,
583
+ "loss": 0.1469,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.26895657809462087,
588
+ "grad_norm": 0.3621589576589696,
589
+ "learning_rate": 4.579753655414916e-06,
590
+ "loss": 0.1469,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.27219701879455604,
595
+ "grad_norm": 0.38775207709700354,
596
+ "learning_rate": 4.563926422991973e-06,
597
+ "loss": 0.1507,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.2754374594944913,
602
+ "grad_norm": 0.3808349919379576,
603
+ "learning_rate": 4.5478350485062275e-06,
604
+ "loss": 0.1505,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.27867790019442645,
609
+ "grad_norm": 0.41092711058175224,
610
+ "learning_rate": 4.531481591337706e-06,
611
+ "loss": 0.1465,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.28191834089436163,
616
+ "grad_norm": 0.3677672974804986,
617
+ "learning_rate": 4.51486814440787e-06,
618
+ "loss": 0.1497,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.2851587815942968,
623
+ "grad_norm": 0.3623506682005845,
624
+ "learning_rate": 4.497996833911771e-06,
625
+ "loss": 0.1494,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.28839922229423204,
630
+ "grad_norm": 0.37070206947839884,
631
+ "learning_rate": 4.48086981904593e-06,
632
+ "loss": 0.1543,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.2916396629941672,
637
+ "grad_norm": 0.3639901651185149,
638
+ "learning_rate": 4.463489291732011e-06,
639
+ "loss": 0.1414,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.2948801036941024,
644
+ "grad_norm": 0.3559919942818163,
645
+ "learning_rate": 4.445857476336292e-06,
646
+ "loss": 0.1486,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.29812054439403757,
651
+ "grad_norm": 0.38283982373110453,
652
+ "learning_rate": 4.427976629384994e-06,
653
+ "loss": 0.1465,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.3013609850939728,
658
+ "grad_norm": 0.37801217199677106,
659
+ "learning_rate": 4.409849039275486e-06,
660
+ "loss": 0.1468,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.304601425793908,
665
+ "grad_norm": 0.4096620960705283,
666
+ "learning_rate": 4.391477025983418e-06,
667
+ "loss": 0.1479,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.30784186649384315,
672
+ "grad_norm": 0.3604056272858565,
673
+ "learning_rate": 4.372862940765811e-06,
674
+ "loss": 0.1496,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.31108230719377833,
679
+ "grad_norm": 0.3538536423133527,
680
+ "learning_rate": 4.354009165860136e-06,
681
+ "loss": 0.1482,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.31432274789371356,
686
+ "grad_norm": 0.3904102444609031,
687
+ "learning_rate": 4.3349181141794414e-06,
688
+ "loss": 0.1435,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.31756318859364874,
693
+ "grad_norm": 0.3859901702462496,
694
+ "learning_rate": 4.315592229003544e-06,
695
+ "loss": 0.1456,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.3208036292935839,
700
+ "grad_norm": 0.35852358880201146,
701
+ "learning_rate": 4.2960339836663364e-06,
702
+ "loss": 0.145,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.32404406999351915,
707
+ "grad_norm": 0.3565242267972036,
708
+ "learning_rate": 4.276245881239253e-06,
709
+ "loss": 0.1466,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.3272845106934543,
714
+ "grad_norm": 0.3433337357670066,
715
+ "learning_rate": 4.2562304542109186e-06,
716
+ "loss": 0.1439,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.3305249513933895,
721
+ "grad_norm": 0.38264147572948926,
722
+ "learning_rate": 4.235990264163048e-06,
723
+ "loss": 0.1442,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.3337653920933247,
728
+ "grad_norm": 0.36500282997002936,
729
+ "learning_rate": 4.215527901442609e-06,
730
+ "loss": 0.1433,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.3370058327932599,
735
+ "grad_norm": 0.3846660553174048,
736
+ "learning_rate": 4.194845984830308e-06,
737
+ "loss": 0.1474,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.3402462734931951,
742
+ "grad_norm": 0.37675710340654855,
743
+ "learning_rate": 4.173947161205443e-06,
744
+ "loss": 0.149,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.34348671419313026,
749
+ "grad_norm": 0.36661580250234177,
750
+ "learning_rate": 4.152834105207146e-06,
751
+ "loss": 0.1448,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.34672715489306544,
756
+ "grad_norm": 0.37636552154818864,
757
+ "learning_rate": 4.131509518892093e-06,
758
+ "loss": 0.1454,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.34996759559300067,
763
+ "grad_norm": 0.3581018731822282,
764
+ "learning_rate": 4.109976131388682e-06,
765
+ "loss": 0.1469,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.35320803629293585,
770
+ "grad_norm": 0.3387824669850879,
771
+ "learning_rate": 4.088236698547767e-06,
772
+ "loss": 0.1453,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.356448476992871,
777
+ "grad_norm": 0.3691931165432145,
778
+ "learning_rate": 4.06629400258996e-06,
779
+ "loss": 0.1466,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.3596889176928062,
784
+ "grad_norm": 0.34829825491346456,
785
+ "learning_rate": 4.0441508517495605e-06,
786
+ "loss": 0.1449,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.36292935839274143,
791
+ "grad_norm": 0.38098751196774483,
792
+ "learning_rate": 4.021810079915159e-06,
793
+ "loss": 0.1466,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.3661697990926766,
798
+ "grad_norm": 0.3742527627607329,
799
+ "learning_rate": 3.9992745462669515e-06,
800
+ "loss": 0.1467,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.3694102397926118,
805
+ "grad_norm": 0.3905850349665511,
806
+ "learning_rate": 3.976547134910826e-06,
807
+ "loss": 0.1452,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.37265068049254696,
812
+ "grad_norm": 0.332721575400441,
813
+ "learning_rate": 3.953630754509245e-06,
814
+ "loss": 0.1478,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.3758911211924822,
819
+ "grad_norm": 0.34472432311127355,
820
+ "learning_rate": 3.930528337909002e-06,
821
+ "loss": 0.1466,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.37913156189241737,
826
+ "grad_norm": 0.3742114947748869,
827
+ "learning_rate": 3.907242841765874e-06,
828
+ "loss": 0.1475,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.38237200259235254,
833
+ "grad_norm": 0.363030396921668,
834
+ "learning_rate": 3.883777246166219e-06,
835
+ "loss": 0.1386,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.3856124432922878,
840
+ "grad_norm": 0.36210921666900014,
841
+ "learning_rate": 3.860134554245597e-06,
842
+ "loss": 0.1412,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.38885288399222295,
847
+ "grad_norm": 0.3599182096747789,
848
+ "learning_rate": 3.836317791804415e-06,
849
+ "loss": 0.1408,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.39209332469215813,
854
+ "grad_norm": 0.40331300021325894,
855
+ "learning_rate": 3.81233000692069e-06,
856
+ "loss": 0.1428,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.3953337653920933,
861
+ "grad_norm": 0.34914858021538103,
862
+ "learning_rate": 3.7881742695599567e-06,
863
+ "loss": 0.1439,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.39857420609202854,
868
+ "grad_norm": 0.3637051905577843,
869
+ "learning_rate": 3.763853671182367e-06,
870
+ "loss": 0.1445,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.4018146467919637,
875
+ "grad_norm": 0.3560295518397911,
876
+ "learning_rate": 3.7393713243470475e-06,
877
+ "loss": 0.1426,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.4050550874918989,
882
+ "grad_norm": 0.37115109403035373,
883
+ "learning_rate": 3.7147303623137508e-06,
884
+ "loss": 0.1438,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.40829552819183407,
889
+ "grad_norm": 0.36713310156131235,
890
+ "learning_rate": 3.6899339386418635e-06,
891
+ "loss": 0.1433,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.4115359688917693,
896
+ "grad_norm": 0.3739701552036908,
897
+ "learning_rate": 3.664985226786807e-06,
898
+ "loss": 0.1415,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.4147764095917045,
903
+ "grad_norm": 0.4018803004726532,
904
+ "learning_rate": 3.6398874196939015e-06,
905
+ "loss": 0.1424,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.41801685029163965,
910
+ "grad_norm": 0.3856228913000613,
911
+ "learning_rate": 3.614643729389731e-06,
912
+ "loss": 0.1434,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.42125729099157483,
917
+ "grad_norm": 0.3590367113459008,
918
+ "learning_rate": 3.5892573865710665e-06,
919
+ "loss": 0.1447,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.42449773169151006,
924
+ "grad_norm": 0.35395438465776946,
925
+ "learning_rate": 3.5637316401913997e-06,
926
+ "loss": 0.1432,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.42773817239144524,
931
+ "grad_norm": 0.35628640020346514,
932
+ "learning_rate": 3.5380697570451413e-06,
933
+ "loss": 0.1404,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.4309786130913804,
938
+ "grad_norm": 0.36552890562756624,
939
+ "learning_rate": 3.5122750213495355e-06,
940
+ "loss": 0.1456,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.43421905379131565,
945
+ "grad_norm": 0.37173141387082576,
946
+ "learning_rate": 3.486350734324346e-06,
947
+ "loss": 0.1435,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.4374594944912508,
952
+ "grad_norm": 0.3539914127253098,
953
+ "learning_rate": 3.460300213769361e-06,
954
+ "loss": 0.1427,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.440699935191186,
959
+ "grad_norm": 0.3619461607830295,
960
+ "learning_rate": 3.4341267936397843e-06,
961
+ "loss": 0.1403,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.4439403758911212,
966
+ "grad_norm": 0.35261727478654553,
967
+ "learning_rate": 3.4078338236195506e-06,
968
+ "loss": 0.1419,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.4471808165910564,
973
+ "grad_norm": 0.3320165163284001,
974
+ "learning_rate": 3.3814246686926367e-06,
975
+ "loss": 0.1422,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.4504212572909916,
980
+ "grad_norm": 0.3505007893913486,
981
+ "learning_rate": 3.354902708712402e-06,
982
+ "loss": 0.1415,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.45366169799092676,
987
+ "grad_norm": 0.366809997754886,
988
+ "learning_rate": 3.3282713379690418e-06,
989
+ "loss": 0.1423,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.45690213869086194,
994
+ "grad_norm": 0.37510389340397243,
995
+ "learning_rate": 3.301533964755179e-06,
996
+ "loss": 0.144,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.46014257939079717,
1001
+ "grad_norm": 0.3751915143045206,
1002
+ "learning_rate": 3.2746940109296744e-06,
1003
+ "loss": 0.143,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.46338302009073234,
1008
+ "grad_norm": 0.34874611917779574,
1009
+ "learning_rate": 3.2477549114796892e-06,
1010
+ "loss": 0.1433,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.4666234607906675,
1015
+ "grad_norm": 0.34401160771032996,
1016
+ "learning_rate": 3.220720114081081e-06,
1017
+ "loss": 0.1429,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.4698639014906027,
1022
+ "grad_norm": 0.34265687524195426,
1023
+ "learning_rate": 3.193593078657165e-06,
1024
+ "loss": 0.1447,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.47310434219053793,
1029
+ "grad_norm": 0.36594432029338536,
1030
+ "learning_rate": 3.1663772769359134e-06,
1031
+ "loss": 0.1417,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.4763447828904731,
1036
+ "grad_norm": 0.4010466659928031,
1037
+ "learning_rate": 3.1390761920056423e-06,
1038
+ "loss": 0.1449,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.4795852235904083,
1043
+ "grad_norm": 0.38421005144265497,
1044
+ "learning_rate": 3.111693317869241e-06,
1045
+ "loss": 0.1415,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.48282566429034346,
1050
+ "grad_norm": 0.3405551521124088,
1051
+ "learning_rate": 3.084232158997017e-06,
1052
+ "loss": 0.1416,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.4860661049902787,
1057
+ "grad_norm": 0.34368244054627933,
1058
+ "learning_rate": 3.0566962298781822e-06,
1059
+ "loss": 0.1366,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.48930654569021387,
1064
+ "grad_norm": 0.3475737652432212,
1065
+ "learning_rate": 3.0290890545710732e-06,
1066
+ "loss": 0.1427,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.49254698639014904,
1071
+ "grad_norm": 0.3915309068687966,
1072
+ "learning_rate": 3.0014141662521423e-06,
1073
+ "loss": 0.142,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.4957874270900843,
1078
+ "grad_norm": 0.38245602033100184,
1079
+ "learning_rate": 2.973675106763776e-06,
1080
+ "loss": 0.1482,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.49902786779001945,
1085
+ "grad_norm": 0.3660718954078658,
1086
+ "learning_rate": 2.9458754261610113e-06,
1087
+ "loss": 0.1382,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.5022683084899546,
1092
+ "grad_norm": 0.3560275044859087,
1093
+ "learning_rate": 2.9180186822571953e-06,
1094
+ "loss": 0.1397,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.5055087491898899,
1099
+ "grad_norm": 0.3476891353623016,
1100
+ "learning_rate": 2.8901084401686635e-06,
1101
+ "loss": 0.1399,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.508749189889825,
1106
+ "grad_norm": 0.36360266266143876,
1107
+ "learning_rate": 2.8621482718584615e-06,
1108
+ "loss": 0.1434,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.5119896305897602,
1113
+ "grad_norm": 0.3607127979390933,
1114
+ "learning_rate": 2.8341417556792163e-06,
1115
+ "loss": 0.1391,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.5152300712896954,
1120
+ "grad_norm": 0.36135212438772696,
1121
+ "learning_rate": 2.8060924759151687e-06,
1122
+ "loss": 0.1369,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.5184705119896306,
1127
+ "grad_norm": 0.36165014897143566,
1128
+ "learning_rate": 2.7780040223234623e-06,
1129
+ "loss": 0.1428,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.5217109526895658,
1134
+ "grad_norm": 0.3809583977124813,
1135
+ "learning_rate": 2.749879989674718e-06,
1136
+ "loss": 0.1422,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.5249513933895009,
1141
+ "grad_norm": 0.3385971069870571,
1142
+ "learning_rate": 2.72172397729298e-06,
1143
+ "loss": 0.1356,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.5281918340894362,
1148
+ "grad_norm": 0.3446592011239239,
1149
+ "learning_rate": 2.6935395885950705e-06,
1150
+ "loss": 0.1421,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.5314322747893714,
1155
+ "grad_norm": 0.3824673660290881,
1156
+ "learning_rate": 2.665330430629421e-06,
1157
+ "loss": 0.1397,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.5346727154893065,
1162
+ "grad_norm": 0.3717938840273103,
1163
+ "learning_rate": 2.6371001136144458e-06,
1164
+ "loss": 0.1388,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.5379131561892417,
1169
+ "grad_norm": 0.3442850639482792,
1170
+ "learning_rate": 2.6088522504764983e-06,
1171
+ "loss": 0.1412,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.541153596889177,
1176
+ "grad_norm": 0.35531304056249186,
1177
+ "learning_rate": 2.580590456387497e-06,
1178
+ "loss": 0.1392,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.5443940375891121,
1183
+ "grad_norm": 0.3659296727759621,
1184
+ "learning_rate": 2.5523183483022406e-06,
1185
+ "loss": 0.142,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.5476344782890473,
1190
+ "grad_norm": 0.3598598578768378,
1191
+ "learning_rate": 2.5240395444955255e-06,
1192
+ "loss": 0.14,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.5508749189889826,
1197
+ "grad_norm": 0.37315624702714517,
1198
+ "learning_rate": 2.495757664099063e-06,
1199
+ "loss": 0.1422,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.5541153596889177,
1204
+ "grad_norm": 0.38147815062716345,
1205
+ "learning_rate": 2.4674763266383096e-06,
1206
+ "loss": 0.1399,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.5573558003888529,
1211
+ "grad_norm": 0.3514332109313319,
1212
+ "learning_rate": 2.439199151569236e-06,
1213
+ "loss": 0.1389,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.560596241088788,
1218
+ "grad_norm": 0.3466932777873452,
1219
+ "learning_rate": 2.41092975781511e-06,
1220
+ "loss": 0.1364,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.5638366817887233,
1225
+ "grad_norm": 0.34717040583506037,
1226
+ "learning_rate": 2.382671763303342e-06,
1227
+ "loss": 0.1356,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.5670771224886585,
1232
+ "grad_norm": 0.37026146298749535,
1233
+ "learning_rate": 2.3544287845024643e-06,
1234
+ "loss": 0.1399,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.5703175631885936,
1239
+ "grad_norm": 0.3785438614418004,
1240
+ "learning_rate": 2.326204435959291e-06,
1241
+ "loss": 0.1442,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.5735580038885288,
1246
+ "grad_norm": 0.3809262572423141,
1247
+ "learning_rate": 2.2980023298363334e-06,
1248
+ "loss": 0.142,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.5767984445884641,
1253
+ "grad_norm": 0.3736610534336298,
1254
+ "learning_rate": 2.2698260754495057e-06,
1255
+ "loss": 0.1386,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.5800388852883992,
1260
+ "grad_norm": 0.37243260617527046,
1261
+ "learning_rate": 2.2416792788062073e-06,
1262
+ "loss": 0.1379,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.5832793259883344,
1267
+ "grad_norm": 0.3653198160876884,
1268
+ "learning_rate": 2.213565542143828e-06,
1269
+ "loss": 0.1394,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.5865197666882696,
1274
+ "grad_norm": 0.3456465195181901,
1275
+ "learning_rate": 2.1854884634687267e-06,
1276
+ "loss": 0.1361,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.5897602073882048,
1281
+ "grad_norm": 0.35698643784051837,
1282
+ "learning_rate": 2.1574516360957588e-06,
1283
+ "loss": 0.134,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.59300064808814,
1288
+ "grad_norm": 0.33396274798278874,
1289
+ "learning_rate": 2.129458648188405e-06,
1290
+ "loss": 0.1378,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.5962410887880751,
1295
+ "grad_norm": 0.35679763524806046,
1296
+ "learning_rate": 2.1015130822995553e-06,
1297
+ "loss": 0.1387,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.5994815294880104,
1302
+ "grad_norm": 0.3748507983654056,
1303
+ "learning_rate": 2.0736185149130117e-06,
1304
+ "loss": 0.1353,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.6027219701879456,
1309
+ "grad_norm": 0.3680336654813566,
1310
+ "learning_rate": 2.0457785159857673e-06,
1311
+ "loss": 0.142,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.6059624108878807,
1316
+ "grad_norm": 0.3461769829954041,
1317
+ "learning_rate": 2.0179966484911268e-06,
1318
+ "loss": 0.1371,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.609202851587816,
1323
+ "grad_norm": 0.35213432555022733,
1324
+ "learning_rate": 1.990276467962711e-06,
1325
+ "loss": 0.1377,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.6124432922877512,
1330
+ "grad_norm": 0.3838660410707078,
1331
+ "learning_rate": 1.9626215220394213e-06,
1332
+ "loss": 0.1403,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.6156837329876863,
1337
+ "grad_norm": 0.394603150983814,
1338
+ "learning_rate": 1.9350353500114072e-06,
1339
+ "loss": 0.1369,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.6189241736876215,
1344
+ "grad_norm": 0.3327444517938039,
1345
+ "learning_rate": 1.907521482367114e-06,
1346
+ "loss": 0.1406,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.6221646143875567,
1351
+ "grad_norm": 0.3704170243923683,
1352
+ "learning_rate": 1.8800834403414437e-06,
1353
+ "loss": 0.137,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.6254050550874919,
1358
+ "grad_norm": 0.40349804036311127,
1359
+ "learning_rate": 1.8527247354651048e-06,
1360
+ "loss": 0.1362,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.6286454957874271,
1365
+ "grad_norm": 0.364607317668594,
1366
+ "learning_rate": 1.8254488691152145e-06,
1367
+ "loss": 0.1378,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.6318859364873622,
1372
+ "grad_norm": 0.38797616750444036,
1373
+ "learning_rate": 1.7982593320671793e-06,
1374
+ "loss": 0.1387,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.6351263771872975,
1379
+ "grad_norm": 0.3943004101702645,
1380
+ "learning_rate": 1.7711596040479523e-06,
1381
+ "loss": 0.1378,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.6383668178872327,
1386
+ "grad_norm": 0.34986766543947917,
1387
+ "learning_rate": 1.744153153290693e-06,
1388
+ "loss": 0.1382,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.6416072585871678,
1393
+ "grad_norm": 0.35689489846966777,
1394
+ "learning_rate": 1.717243436090904e-06,
1395
+ "loss": 0.1398,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.6448476992871031,
1400
+ "grad_norm": 0.358287736012017,
1401
+ "learning_rate": 1.6904338963640913e-06,
1402
+ "loss": 0.1342,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.6480881399870383,
1407
+ "grad_norm": 0.35009264406089535,
1408
+ "learning_rate": 1.6637279652050103e-06,
1409
+ "loss": 0.1384,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.6513285806869734,
1414
+ "grad_norm": 0.36264463552008447,
1415
+ "learning_rate": 1.6371290604485532e-06,
1416
+ "loss": 0.1345,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.6545690213869086,
1421
+ "grad_norm": 0.37898011262102793,
1422
+ "learning_rate": 1.610640586232336e-06,
1423
+ "loss": 0.136,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.6578094620868438,
1428
+ "grad_norm": 0.3486402988937132,
1429
+ "learning_rate": 1.5842659325610326e-06,
1430
+ "loss": 0.1339,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.661049902786779,
1435
+ "grad_norm": 0.3494806553044058,
1436
+ "learning_rate": 1.5580084748725193e-06,
1437
+ "loss": 0.132,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 0.6642903434867142,
1442
+ "grad_norm": 0.3376036564372793,
1443
+ "learning_rate": 1.5318715736058926e-06,
1444
+ "loss": 0.1419,
1445
+ "step": 2050
1446
+ },
1447
+ {
1448
+ "epoch": 0.6675307841866494,
1449
+ "grad_norm": 0.3619041141435384,
1450
+ "learning_rate": 1.5058585737713883e-06,
1451
+ "loss": 0.1377,
1452
+ "step": 2060
1453
+ },
1454
+ {
1455
+ "epoch": 0.6707712248865846,
1456
+ "grad_norm": 0.37485938510309685,
1457
+ "learning_rate": 1.4799728045222956e-06,
1458
+ "loss": 0.1373,
1459
+ "step": 2070
1460
+ },
1461
+ {
1462
+ "epoch": 0.6740116655865198,
1463
+ "grad_norm": 0.3596734953623062,
1464
+ "learning_rate": 1.4542175787288818e-06,
1465
+ "loss": 0.1396,
1466
+ "step": 2080
1467
+ },
1468
+ {
1469
+ "epoch": 0.6772521062864549,
1470
+ "grad_norm": 0.3531801559136217,
1471
+ "learning_rate": 1.428596192554421e-06,
1472
+ "loss": 0.139,
1473
+ "step": 2090
1474
+ },
1475
+ {
1476
+ "epoch": 0.6804925469863902,
1477
+ "grad_norm": 0.35403583546745904,
1478
+ "learning_rate": 1.4031119250333393e-06,
1479
+ "loss": 0.1386,
1480
+ "step": 2100
1481
+ },
1482
+ {
1483
+ "epoch": 0.6837329876863253,
1484
+ "grad_norm": 0.3499312200329266,
1485
+ "learning_rate": 1.3777680376515667e-06,
1486
+ "loss": 0.138,
1487
+ "step": 2110
1488
+ },
1489
+ {
1490
+ "epoch": 0.6869734283862605,
1491
+ "grad_norm": 0.3655179914285288,
1492
+ "learning_rate": 1.3525677739291338e-06,
1493
+ "loss": 0.1374,
1494
+ "step": 2120
1495
+ },
1496
+ {
1497
+ "epoch": 0.6902138690861958,
1498
+ "grad_norm": 0.36296013901637053,
1499
+ "learning_rate": 1.3275143590050622e-06,
1500
+ "loss": 0.1391,
1501
+ "step": 2130
1502
+ },
1503
+ {
1504
+ "epoch": 0.6934543097861309,
1505
+ "grad_norm": 0.33352013543504444,
1506
+ "learning_rate": 1.3026109992246084e-06,
1507
+ "loss": 0.1356,
1508
+ "step": 2140
1509
+ },
1510
+ {
1511
+ "epoch": 0.6966947504860661,
1512
+ "grad_norm": 0.34615241767967,
1513
+ "learning_rate": 1.2778608817289167e-06,
1514
+ "loss": 0.1365,
1515
+ "step": 2150
1516
+ },
1517
+ {
1518
+ "epoch": 0.6999351911860013,
1519
+ "grad_norm": 0.36999888059560476,
1520
+ "learning_rate": 1.2532671740471292e-06,
1521
+ "loss": 0.1411,
1522
+ "step": 2160
1523
+ },
1524
+ {
1525
+ "epoch": 0.7031756318859365,
1526
+ "grad_norm": 0.34402496133806365,
1527
+ "learning_rate": 1.2288330236910025e-06,
1528
+ "loss": 0.1396,
1529
+ "step": 2170
1530
+ },
1531
+ {
1532
+ "epoch": 0.7064160725858717,
1533
+ "grad_norm": 0.35636630900488314,
1534
+ "learning_rate": 1.204561557752085e-06,
1535
+ "loss": 0.1394,
1536
+ "step": 2180
1537
+ },
1538
+ {
1539
+ "epoch": 0.7096565132858069,
1540
+ "grad_norm": 0.36181518745129104,
1541
+ "learning_rate": 1.1804558825015172e-06,
1542
+ "loss": 0.1365,
1543
+ "step": 2190
1544
+ },
1545
+ {
1546
+ "epoch": 0.712896953985742,
1547
+ "grad_norm": 0.3563465165177732,
1548
+ "learning_rate": 1.1565190829924838e-06,
1549
+ "loss": 0.1362,
1550
+ "step": 2200
1551
+ },
1552
+ {
1553
+ "epoch": 0.7161373946856773,
1554
+ "grad_norm": 0.33363782347377735,
1555
+ "learning_rate": 1.13275422266539e-06,
1556
+ "loss": 0.1398,
1557
+ "step": 2210
1558
+ },
1559
+ {
1560
+ "epoch": 0.7193778353856124,
1561
+ "grad_norm": 0.36414316020816895,
1562
+ "learning_rate": 1.109164342955803e-06,
1563
+ "loss": 0.1383,
1564
+ "step": 2220
1565
+ },
1566
+ {
1567
+ "epoch": 0.7226182760855476,
1568
+ "grad_norm": 0.36241713354650357,
1569
+ "learning_rate": 1.0857524629052057e-06,
1570
+ "loss": 0.1373,
1571
+ "step": 2230
1572
+ },
1573
+ {
1574
+ "epoch": 0.7258587167854829,
1575
+ "grad_norm": 0.35575126832820747,
1576
+ "learning_rate": 1.0625215787746185e-06,
1577
+ "loss": 0.1376,
1578
+ "step": 2240
1579
+ },
1580
+ {
1581
+ "epoch": 0.729099157485418,
1582
+ "grad_norm": 0.3581511044472468,
1583
+ "learning_rate": 1.0394746636611364e-06,
1584
+ "loss": 0.1327,
1585
+ "step": 2250
1586
+ },
1587
+ {
1588
+ "epoch": 0.7323395981853532,
1589
+ "grad_norm": 0.3487467990800773,
1590
+ "learning_rate": 1.016614667117434e-06,
1591
+ "loss": 0.1362,
1592
+ "step": 2260
1593
+ },
1594
+ {
1595
+ "epoch": 0.7355800388852884,
1596
+ "grad_norm": 0.36540009173824517,
1597
+ "learning_rate": 9.939445147742786e-07,
1598
+ "loss": 0.1408,
1599
+ "step": 2270
1600
+ },
1601
+ {
1602
+ "epoch": 0.7388204795852236,
1603
+ "grad_norm": 0.3504860844231358,
1604
+ "learning_rate": 9.714671079661048e-07,
1605
+ "loss": 0.1383,
1606
+ "step": 2280
1607
+ },
1608
+ {
1609
+ "epoch": 0.7420609202851588,
1610
+ "grad_norm": 0.3546645739584433,
1611
+ "learning_rate": 9.491853233597037e-07,
1612
+ "loss": 0.1367,
1613
+ "step": 2290
1614
+ },
1615
+ {
1616
+ "epoch": 0.7453013609850939,
1617
+ "grad_norm": 0.36305507134438186,
1618
+ "learning_rate": 9.271020125860658e-07,
1619
+ "loss": 0.1367,
1620
+ "step": 2300
1621
+ },
1622
+ {
1623
+ "epoch": 0.7485418016850292,
1624
+ "grad_norm": 0.3632120141566518,
1625
+ "learning_rate": 9.052200018754276e-07,
1626
+ "loss": 0.1364,
1627
+ "step": 2310
1628
+ },
1629
+ {
1630
+ "epoch": 0.7517822423849644,
1631
+ "grad_norm": 0.3222399957976063,
1632
+ "learning_rate": 8.835420916955686e-07,
1633
+ "loss": 0.1376,
1634
+ "step": 2320
1635
+ },
1636
+ {
1637
+ "epoch": 0.7550226830848995,
1638
+ "grad_norm": 0.3648037126049122,
1639
+ "learning_rate": 8.620710563934063e-07,
1640
+ "loss": 0.1374,
1641
+ "step": 2330
1642
+ },
1643
+ {
1644
+ "epoch": 0.7582631237848347,
1645
+ "grad_norm": 0.38493254190484144,
1646
+ "learning_rate": 8.408096438399371e-07,
1647
+ "loss": 0.1369,
1648
+ "step": 2340
1649
+ },
1650
+ {
1651
+ "epoch": 0.76150356448477,
1652
+ "grad_norm": 0.335337634563654,
1653
+ "learning_rate": 8.19760575078562e-07,
1654
+ "loss": 0.1353,
1655
+ "step": 2350
1656
+ },
1657
+ {
1658
+ "epoch": 0.7647440051847051,
1659
+ "grad_norm": 0.34948986108995744,
1660
+ "learning_rate": 7.989265439768412e-07,
1661
+ "loss": 0.1336,
1662
+ "step": 2360
1663
+ },
1664
+ {
1665
+ "epoch": 0.7679844458846403,
1666
+ "grad_norm": 0.3533447598580161,
1667
+ "learning_rate": 7.783102168817406e-07,
1668
+ "loss": 0.1338,
1669
+ "step": 2370
1670
+ },
1671
+ {
1672
+ "epoch": 0.7712248865845756,
1673
+ "grad_norm": 0.35564500184932724,
1674
+ "learning_rate": 7.579142322783837e-07,
1675
+ "loss": 0.1371,
1676
+ "step": 2380
1677
+ },
1678
+ {
1679
+ "epoch": 0.7744653272845107,
1680
+ "grad_norm": 0.3565193952902089,
1681
+ "learning_rate": 7.37741200452384e-07,
1682
+ "loss": 0.1333,
1683
+ "step": 2390
1684
+ },
1685
+ {
1686
+ "epoch": 0.7777057679844459,
1687
+ "grad_norm": 0.38945562323117644,
1688
+ "learning_rate": 7.177937031557733e-07,
1689
+ "loss": 0.1384,
1690
+ "step": 2400
1691
+ },
1692
+ {
1693
+ "epoch": 0.780946208684381,
1694
+ "grad_norm": 0.33188297048475185,
1695
+ "learning_rate": 6.98074293276595e-07,
1696
+ "loss": 0.1304,
1697
+ "step": 2410
1698
+ },
1699
+ {
1700
+ "epoch": 0.7841866493843163,
1701
+ "grad_norm": 0.33600376376796814,
1702
+ "learning_rate": 6.785854945121773e-07,
1703
+ "loss": 0.1371,
1704
+ "step": 2420
1705
+ },
1706
+ {
1707
+ "epoch": 0.7874270900842515,
1708
+ "grad_norm": 0.3733419224635097,
1709
+ "learning_rate": 6.593298010461541e-07,
1710
+ "loss": 0.1357,
1711
+ "step": 2430
1712
+ },
1713
+ {
1714
+ "epoch": 0.7906675307841866,
1715
+ "grad_norm": 0.360994575329127,
1716
+ "learning_rate": 6.403096772292611e-07,
1717
+ "loss": 0.1388,
1718
+ "step": 2440
1719
+ },
1720
+ {
1721
+ "epoch": 0.7939079714841218,
1722
+ "grad_norm": 0.3540220491931335,
1723
+ "learning_rate": 6.215275572639387e-07,
1724
+ "loss": 0.1383,
1725
+ "step": 2450
1726
+ },
1727
+ {
1728
+ "epoch": 0.7971484121840571,
1729
+ "grad_norm": 0.3421614423791235,
1730
+ "learning_rate": 6.029858448928091e-07,
1731
+ "loss": 0.1374,
1732
+ "step": 2460
1733
+ },
1734
+ {
1735
+ "epoch": 0.8003888528839922,
1736
+ "grad_norm": 0.352474503469002,
1737
+ "learning_rate": 5.846869130910413e-07,
1738
+ "loss": 0.1394,
1739
+ "step": 2470
1740
+ },
1741
+ {
1742
+ "epoch": 0.8036292935839274,
1743
+ "grad_norm": 0.33788431994154117,
1744
+ "learning_rate": 5.666331037626585e-07,
1745
+ "loss": 0.1345,
1746
+ "step": 2480
1747
+ },
1748
+ {
1749
+ "epoch": 0.8068697342838627,
1750
+ "grad_norm": 0.32025075334711434,
1751
+ "learning_rate": 5.488267274408207e-07,
1752
+ "loss": 0.1349,
1753
+ "step": 2490
1754
+ },
1755
+ {
1756
+ "epoch": 0.8101101749837978,
1757
+ "grad_norm": 0.36081495813728437,
1758
+ "learning_rate": 5.312700629921181e-07,
1759
+ "loss": 0.1325,
1760
+ "step": 2500
1761
+ },
1762
+ {
1763
+ "epoch": 0.813350615683733,
1764
+ "grad_norm": 0.3416845659191573,
1765
+ "learning_rate": 5.139653573249253e-07,
1766
+ "loss": 0.1333,
1767
+ "step": 2510
1768
+ },
1769
+ {
1770
+ "epoch": 0.8165910563836681,
1771
+ "grad_norm": 0.3831793796512933,
1772
+ "learning_rate": 4.969148251018385e-07,
1773
+ "loss": 0.1342,
1774
+ "step": 2520
1775
+ },
1776
+ {
1777
+ "epoch": 0.8198314970836034,
1778
+ "grad_norm": 0.3578750131718832,
1779
+ "learning_rate": 4.801206484562446e-07,
1780
+ "loss": 0.1352,
1781
+ "step": 2530
1782
+ },
1783
+ {
1784
+ "epoch": 0.8230719377835386,
1785
+ "grad_norm": 0.355502152277739,
1786
+ "learning_rate": 4.6358497671304623e-07,
1787
+ "loss": 0.1357,
1788
+ "step": 2540
1789
+ },
1790
+ {
1791
+ "epoch": 0.8263123784834737,
1792
+ "grad_norm": 0.3510295514913444,
1793
+ "learning_rate": 4.473099261135977e-07,
1794
+ "loss": 0.1388,
1795
+ "step": 2550
1796
+ },
1797
+ {
1798
+ "epoch": 0.829552819183409,
1799
+ "grad_norm": 0.32078925808882724,
1800
+ "learning_rate": 4.312975795448593e-07,
1801
+ "loss": 0.1364,
1802
+ "step": 2560
1803
+ },
1804
+ {
1805
+ "epoch": 0.8327932598833442,
1806
+ "grad_norm": 0.3692542497286855,
1807
+ "learning_rate": 4.1554998627283655e-07,
1808
+ "loss": 0.1315,
1809
+ "step": 2570
1810
+ },
1811
+ {
1812
+ "epoch": 0.8360337005832793,
1813
+ "grad_norm": 0.37137933087389124,
1814
+ "learning_rate": 4.000691616803076e-07,
1815
+ "loss": 0.1413,
1816
+ "step": 2580
1817
+ },
1818
+ {
1819
+ "epoch": 0.8392741412832145,
1820
+ "grad_norm": 0.3882652809375332,
1821
+ "learning_rate": 3.8485708700889913e-07,
1822
+ "loss": 0.1364,
1823
+ "step": 2590
1824
+ },
1825
+ {
1826
+ "epoch": 0.8425145819831497,
1827
+ "grad_norm": 0.36297859765424056,
1828
+ "learning_rate": 3.699157091055225e-07,
1829
+ "loss": 0.1346,
1830
+ "step": 2600
1831
+ },
1832
+ {
1833
+ "epoch": 0.8457550226830849,
1834
+ "grad_norm": 0.3565105681283779,
1835
+ "learning_rate": 3.552469401732181e-07,
1836
+ "loss": 0.1379,
1837
+ "step": 2610
1838
+ },
1839
+ {
1840
+ "epoch": 0.8489954633830201,
1841
+ "grad_norm": 0.33430669724867546,
1842
+ "learning_rate": 3.408526575264298e-07,
1843
+ "loss": 0.1407,
1844
+ "step": 2620
1845
+ },
1846
+ {
1847
+ "epoch": 0.8522359040829552,
1848
+ "grad_norm": 0.3705327571148099,
1849
+ "learning_rate": 3.267347033507456e-07,
1850
+ "loss": 0.1341,
1851
+ "step": 2630
1852
+ },
1853
+ {
1854
+ "epoch": 0.8554763447828905,
1855
+ "grad_norm": 0.3473212789683562,
1856
+ "learning_rate": 3.128948844671334e-07,
1857
+ "loss": 0.1354,
1858
+ "step": 2640
1859
+ },
1860
+ {
1861
+ "epoch": 0.8587167854828257,
1862
+ "grad_norm": 0.35963054694538577,
1863
+ "learning_rate": 2.993349721007041e-07,
1864
+ "loss": 0.136,
1865
+ "step": 2650
1866
+ },
1867
+ {
1868
+ "epoch": 0.8619572261827608,
1869
+ "grad_norm": 0.37555057368085243,
1870
+ "learning_rate": 2.860567016540322e-07,
1871
+ "loss": 0.1379,
1872
+ "step": 2660
1873
+ },
1874
+ {
1875
+ "epoch": 0.8651976668826961,
1876
+ "grad_norm": 0.38221465588612524,
1877
+ "learning_rate": 2.730617724850523e-07,
1878
+ "loss": 0.138,
1879
+ "step": 2670
1880
+ },
1881
+ {
1882
+ "epoch": 0.8684381075826313,
1883
+ "grad_norm": 0.39323307490910586,
1884
+ "learning_rate": 2.603518476895794e-07,
1885
+ "loss": 0.1372,
1886
+ "step": 2680
1887
+ },
1888
+ {
1889
+ "epoch": 0.8716785482825664,
1890
+ "grad_norm": 0.32668418769533897,
1891
+ "learning_rate": 2.479285538884649e-07,
1892
+ "loss": 0.1315,
1893
+ "step": 2690
1894
+ },
1895
+ {
1896
+ "epoch": 0.8749189889825016,
1897
+ "grad_norm": 0.36967939498979446,
1898
+ "learning_rate": 2.3579348101941817e-07,
1899
+ "loss": 0.1395,
1900
+ "step": 2700
1901
+ },
1902
+ {
1903
+ "epoch": 0.8781594296824368,
1904
+ "grad_norm": 0.4148617877855533,
1905
+ "learning_rate": 2.239481821335296e-07,
1906
+ "loss": 0.1346,
1907
+ "step": 2710
1908
+ },
1909
+ {
1910
+ "epoch": 0.881399870382372,
1911
+ "grad_norm": 0.3507803299933488,
1912
+ "learning_rate": 2.1239417319650707e-07,
1913
+ "loss": 0.135,
1914
+ "step": 2720
1915
+ },
1916
+ {
1917
+ "epoch": 0.8846403110823072,
1918
+ "grad_norm": 0.3431423389326874,
1919
+ "learning_rate": 2.0113293289466484e-07,
1920
+ "loss": 0.1386,
1921
+ "step": 2730
1922
+ },
1923
+ {
1924
+ "epoch": 0.8878807517822424,
1925
+ "grad_norm": 0.3556269315719522,
1926
+ "learning_rate": 1.901659024456784e-07,
1927
+ "loss": 0.1374,
1928
+ "step": 2740
1929
+ },
1930
+ {
1931
+ "epoch": 0.8911211924821776,
1932
+ "grad_norm": 0.3581655711093919,
1933
+ "learning_rate": 1.7949448541413923e-07,
1934
+ "loss": 0.1349,
1935
+ "step": 2750
1936
+ },
1937
+ {
1938
+ "epoch": 0.8943616331821128,
1939
+ "grad_norm": 0.3539219505292951,
1940
+ "learning_rate": 1.6912004753192463e-07,
1941
+ "loss": 0.1372,
1942
+ "step": 2760
1943
+ },
1944
+ {
1945
+ "epoch": 0.8976020738820479,
1946
+ "grad_norm": 0.36487534167481467,
1947
+ "learning_rate": 1.5904391652341034e-07,
1948
+ "loss": 0.1398,
1949
+ "step": 2770
1950
+ },
1951
+ {
1952
+ "epoch": 0.9008425145819832,
1953
+ "grad_norm": 0.3547252654216283,
1954
+ "learning_rate": 1.492673819355489e-07,
1955
+ "loss": 0.137,
1956
+ "step": 2780
1957
+ },
1958
+ {
1959
+ "epoch": 0.9040829552819183,
1960
+ "grad_norm": 0.3624466589823807,
1961
+ "learning_rate": 1.3979169497283369e-07,
1962
+ "loss": 0.1344,
1963
+ "step": 2790
1964
+ },
1965
+ {
1966
+ "epoch": 0.9073233959818535,
1967
+ "grad_norm": 0.3597593530124065,
1968
+ "learning_rate": 1.3061806833716773e-07,
1969
+ "loss": 0.1351,
1970
+ "step": 2800
1971
+ },
1972
+ {
1973
+ "epoch": 0.9105638366817888,
1974
+ "grad_norm": 0.36978575366992134,
1975
+ "learning_rate": 1.2174767607266214e-07,
1976
+ "loss": 0.1381,
1977
+ "step": 2810
1978
+ },
1979
+ {
1980
+ "epoch": 0.9138042773817239,
1981
+ "grad_norm": 0.3394271179919748,
1982
+ "learning_rate": 1.1318165341538096e-07,
1983
+ "loss": 0.1409,
1984
+ "step": 2820
1985
+ },
1986
+ {
1987
+ "epoch": 0.9170447180816591,
1988
+ "grad_norm": 0.3886245251582624,
1989
+ "learning_rate": 1.0492109664805572e-07,
1990
+ "loss": 0.1357,
1991
+ "step": 2830
1992
+ },
1993
+ {
1994
+ "epoch": 0.9202851587815943,
1995
+ "grad_norm": 0.357080659087376,
1996
+ "learning_rate": 9.69670629597802e-08,
1997
+ "loss": 0.1367,
1998
+ "step": 2840
1999
+ },
2000
+ {
2001
+ "epoch": 0.9235255994815295,
2002
+ "grad_norm": 0.3725663246448965,
2003
+ "learning_rate": 8.932057031071113e-08,
2004
+ "loss": 0.1367,
2005
+ "step": 2850
2006
+ },
2007
+ {
2008
+ "epoch": 0.9267660401814647,
2009
+ "grad_norm": 0.3534059663914867,
2010
+ "learning_rate": 8.198259730178987e-08,
2011
+ "loss": 0.1372,
2012
+ "step": 2860
2013
+ },
2014
+ {
2015
+ "epoch": 0.9300064808813999,
2016
+ "grad_norm": 0.38942343747420805,
2017
+ "learning_rate": 7.495408304950213e-08,
2018
+ "loss": 0.1362,
2019
+ "step": 2870
2020
+ },
2021
+ {
2022
+ "epoch": 0.933246921581335,
2023
+ "grad_norm": 0.3482911345014878,
2024
+ "learning_rate": 6.823592706568738e-08,
2025
+ "loss": 0.1364,
2026
+ "step": 2880
2027
+ },
2028
+ {
2029
+ "epoch": 0.9364873622812703,
2030
+ "grad_norm": 0.3962969062391369,
2031
+ "learning_rate": 6.182898914241897e-08,
2032
+ "loss": 0.1397,
2033
+ "step": 2890
2034
+ },
2035
+ {
2036
+ "epoch": 0.9397278029812054,
2037
+ "grad_norm": 0.3486632243006723,
2038
+ "learning_rate": 5.5734089241969434e-08,
2039
+ "loss": 0.1355,
2040
+ "step": 2900
2041
+ },
2042
+ {
2043
+ "epoch": 0.9429682436811406,
2044
+ "grad_norm": 0.3414359245355013,
2045
+ "learning_rate": 4.995200739186967e-08,
2046
+ "loss": 0.1353,
2047
+ "step": 2910
2048
+ },
2049
+ {
2050
+ "epoch": 0.9462086843810759,
2051
+ "grad_norm": 0.3577826479652381,
2052
+ "learning_rate": 4.448348358508187e-08,
2053
+ "loss": 0.1349,
2054
+ "step": 2920
2055
+ },
2056
+ {
2057
+ "epoch": 0.949449125081011,
2058
+ "grad_norm": 0.39657239893518126,
2059
+ "learning_rate": 3.932921768529413e-08,
2060
+ "loss": 0.1391,
2061
+ "step": 2930
2062
+ },
2063
+ {
2064
+ "epoch": 0.9526895657809462,
2065
+ "grad_norm": 0.34868268785953926,
2066
+ "learning_rate": 3.448986933735299e-08,
2067
+ "loss": 0.1408,
2068
+ "step": 2940
2069
+ },
2070
+ {
2071
+ "epoch": 0.9559300064808814,
2072
+ "grad_norm": 0.37214717947029213,
2073
+ "learning_rate": 2.996605788283924e-08,
2074
+ "loss": 0.1376,
2075
+ "step": 2950
2076
+ },
2077
+ {
2078
+ "epoch": 0.9591704471808166,
2079
+ "grad_norm": 0.37280976502774077,
2080
+ "learning_rate": 2.5758362280807192e-08,
2081
+ "loss": 0.1323,
2082
+ "step": 2960
2083
+ },
2084
+ {
2085
+ "epoch": 0.9624108878807518,
2086
+ "grad_norm": 0.3529393622845644,
2087
+ "learning_rate": 2.1867321033687827e-08,
2088
+ "loss": 0.1349,
2089
+ "step": 2970
2090
+ },
2091
+ {
2092
+ "epoch": 0.9656513285806869,
2093
+ "grad_norm": 0.344394233275576,
2094
+ "learning_rate": 1.8293432118371146e-08,
2095
+ "loss": 0.1358,
2096
+ "step": 2980
2097
+ },
2098
+ {
2099
+ "epoch": 0.9688917692806222,
2100
+ "grad_norm": 0.3857968624258438,
2101
+ "learning_rate": 1.5037152922475207e-08,
2102
+ "loss": 0.1406,
2103
+ "step": 2990
2104
+ },
2105
+ {
2106
+ "epoch": 0.9721322099805574,
2107
+ "grad_norm": 0.37594692090340304,
2108
+ "learning_rate": 1.2098900185808781e-08,
2109
+ "loss": 0.1362,
2110
+ "step": 3000
2111
+ },
2112
+ {
2113
+ "epoch": 0.9753726506804925,
2114
+ "grad_norm": 0.345262397252086,
2115
+ "learning_rate": 9.479049947038743e-09,
2116
+ "loss": 0.1355,
2117
+ "step": 3010
2118
+ },
2119
+ {
2120
+ "epoch": 0.9786130913804277,
2121
+ "grad_norm": 0.39601712390863364,
2122
+ "learning_rate": 7.177937495561616e-09,
2123
+ "loss": 0.134,
2124
+ "step": 3020
2125
+ },
2126
+ {
2127
+ "epoch": 0.981853532080363,
2128
+ "grad_norm": 0.3544450146581611,
2129
+ "learning_rate": 5.1958573285959546e-09,
2130
+ "loss": 0.1367,
2131
+ "step": 3030
2132
+ },
2133
+ {
2134
+ "epoch": 0.9850939727802981,
2135
+ "grad_norm": 0.36321038651711207,
2136
+ "learning_rate": 3.533063113490276e-09,
2137
+ "loss": 0.1353,
2138
+ "step": 3040
2139
+ },
2140
+ {
2141
+ "epoch": 0.9883344134802333,
2142
+ "grad_norm": 0.34666506280478593,
2143
+ "learning_rate": 2.189767655259578e-09,
2144
+ "loss": 0.132,
2145
+ "step": 3050
2146
+ },
2147
+ {
2148
+ "epoch": 0.9915748541801686,
2149
+ "grad_norm": 0.34052736841522746,
2150
+ "learning_rate": 1.1661428693512944e-09,
2151
+ "loss": 0.1306,
2152
+ "step": 3060
2153
+ },
2154
+ {
2155
+ "epoch": 0.9948152948801037,
2156
+ "grad_norm": 0.31547647192990236,
2157
+ "learning_rate": 4.623197596417828e-10,
2158
+ "loss": 0.1314,
2159
+ "step": 3070
2160
+ },
2161
+ {
2162
+ "epoch": 0.9980557355800389,
2163
+ "grad_norm": 0.3819807577926737,
2164
+ "learning_rate": 7.838840167168071e-11,
2165
+ "loss": 0.1368,
2166
+ "step": 3080
2167
+ },
2168
+ {
2169
+ "epoch": 1.0,
2170
+ "step": 3086,
2171
+ "total_flos": 190452489388032.0,
2172
+ "train_loss": 0.15608682801641588,
2173
+ "train_runtime": 8265.837,
2174
+ "train_samples_per_second": 47.787,
2175
+ "train_steps_per_second": 0.373
2176
+ }
2177
+ ],
2178
+ "logging_steps": 10,
2179
+ "max_steps": 3086,
2180
+ "num_input_tokens_seen": 0,
2181
+ "num_train_epochs": 1,
2182
+ "save_steps": 5000,
2183
+ "stateful_callbacks": {
2184
+ "TrainerControl": {
2185
+ "args": {
2186
+ "should_epoch_stop": false,
2187
+ "should_evaluate": false,
2188
+ "should_log": false,
2189
+ "should_save": true,
2190
+ "should_training_stop": true
2191
+ },
2192
+ "attributes": {}
2193
+ }
2194
+ },
2195
+ "total_flos": 190452489388032.0,
2196
+ "train_batch_size": 16,
2197
+ "trial_name": null,
2198
+ "trial_params": null
2199
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83d23c92d9169c2cc44d83f2a0eb2470f81ae5acda2578a43b444dccf5541a88
3
+ size 7953
training_loss.png ADDED
vocab.json ADDED
The diff for this file is too large to render. See raw diff