Hugging Face's logo

allura-forge
/
q3-30b-ft-fsdp2-adpt-weights-ep1

PEFT
qwen3_moe
Model card Files
xet
Community
Fizzarolli commited on
Commit
a2c313b
·
verified ·
1 Parent(s): a78f9f4

Upload folder using huggingface_hub

Browse files
Files changed (38) hide show
  1. .gitattributes +19 -0
  2. README.md +202 -0
  3. adapter_config.json +40 -0
  4. added_tokens.json +28 -0
  5. config.json +38 -0
  6. merges.txt +0 -0
  7. optimizer_0/.metadata +3 -0
  8. optimizer_0/__0_0.distcp +3 -0
  9. optimizer_0/__1_0.distcp +3 -0
  10. optimizer_0/__2_0.distcp +3 -0
  11. optimizer_0/__3_0.distcp +3 -0
  12. optimizer_0/__4_0.distcp +3 -0
  13. optimizer_0/__5_0.distcp +3 -0
  14. optimizer_0/__6_0.distcp +3 -0
  15. optimizer_0/__7_0.distcp +3 -0
  16. pytorch_model_fsdp_0/.metadata +3 -0
  17. pytorch_model_fsdp_0/__0_0.distcp +3 -0
  18. pytorch_model_fsdp_0/__1_0.distcp +3 -0
  19. pytorch_model_fsdp_0/__2_0.distcp +3 -0
  20. pytorch_model_fsdp_0/__3_0.distcp +3 -0
  21. pytorch_model_fsdp_0/__4_0.distcp +3 -0
  22. pytorch_model_fsdp_0/__5_0.distcp +3 -0
  23. pytorch_model_fsdp_0/__6_0.distcp +3 -0
  24. pytorch_model_fsdp_0/__7_0.distcp +3 -0
  25. rng_state_0.pth +3 -0
  26. rng_state_1.pth +3 -0
  27. rng_state_2.pth +3 -0
  28. rng_state_3.pth +3 -0
  29. rng_state_4.pth +3 -0
  30. rng_state_5.pth +3 -0
  31. rng_state_6.pth +3 -0
  32. rng_state_7.pth +3 -0
  33. scheduler.pt +3 -0
  34. special_tokens_map.json +31 -0
  35. tokenizer.json +3 -0
  36. tokenizer_config.json +240 -0
  37. trainer_state.json +2764 -0
  38. vocab.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,22 @@ 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
+ optimizer_0/.metadata filter=lfs diff=lfs merge=lfs -text
37
+ optimizer_0/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
38
+ optimizer_0/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
39
+ optimizer_0/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
40
+ optimizer_0/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
41
+ optimizer_0/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
42
+ optimizer_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
43
+ optimizer_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
44
+ optimizer_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
45
+ pytorch_model_fsdp_0/.metadata filter=lfs diff=lfs merge=lfs -text
46
+ pytorch_model_fsdp_0/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
47
+ pytorch_model_fsdp_0/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
48
+ pytorch_model_fsdp_0/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
49
+ pytorch_model_fsdp_0/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
50
+ pytorch_model_fsdp_0/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
51
+ pytorch_model_fsdp_0/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
52
+ pytorch_model_fsdp_0/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
53
+ pytorch_model_fsdp_0/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
54
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen3-30B-A3B-Base
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.15.2
adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Qwen/Qwen3-30B-A3B-Base",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": null,
10
+ "inference_mode": false,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 16,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0.35,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "r": 128,
24
+ "rank_pattern": {},
25
+ "revision": null,
26
+ "target_modules": [
27
+ "q_proj",
28
+ "o_proj",
29
+ "k_proj",
30
+ "gate",
31
+ "up_proj",
32
+ "down_proj",
33
+ "gate_proj",
34
+ "v_proj"
35
+ ],
36
+ "task_type": "CAUSAL_LM",
37
+ "trainable_token_indices": null,
38
+ "use_dora": false,
39
+ "use_rslora": true
40
+ }
added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "Qwen3MoeForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "decoder_sparse_step": 1,
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 6144,
15
+ "max_position_embeddings": 32768,
16
+ "max_window_layers": 48,
17
+ "mlp_only_layers": [],
18
+ "model_type": "qwen3_moe",
19
+ "moe_intermediate_size": 768,
20
+ "norm_topk_prob": true,
21
+ "num_attention_heads": 32,
22
+ "num_experts": 128,
23
+ "num_experts_per_tok": 8,
24
+ "num_hidden_layers": 48,
25
+ "num_key_value_heads": 4,
26
+ "output_router_logits": false,
27
+ "rms_norm_eps": 1e-06,
28
+ "rope_scaling": null,
29
+ "rope_theta": 1000000.0,
30
+ "router_aux_loss_coef": 0.001,
31
+ "sliding_window": null,
32
+ "tie_word_embeddings": false,
33
+ "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.51.3",
35
+ "use_cache": false,
36
+ "use_sliding_window": false,
37
+ "vocab_size": 151936
38
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer_0/.metadata ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df129b3ae139dfc2f0c42e8b61764417720e8857d4ed632ad13ad926a5db3b08
3
+ size 114174545
optimizer_0/__0_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a08c5641381ed199d04e1c5db14bc63cb8c4ec102847ebbdd601db46dbb7ba5d
3
+ size 6881886185
optimizer_0/__1_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30e6ee14b77ae9e57404c0893a20d71570b1d38a4e89983f3d5c7d4b69a40703
3
+ size 6890301686
optimizer_0/__2_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7b28bd9d4afbf94e4597c8956d981a46244843e659f34ec8d06d07322ec4a73
3
+ size 6890301686
optimizer_0/__3_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89e5b947e366f00c2decf507553311381c93b246613b2995cc7828a36a5dcec7
3
+ size 6890301686
optimizer_0/__4_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89ca1311ed603e587ff97b60d0487dffe5ecf344980125793e1fcbdcb6ec368e
3
+ size 6890301686
optimizer_0/__5_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07a6e2a0bd0f9a0ce3d7df613f4f1eb46cd5b366021a7872d100159fb9aa9bb6
3
+ size 6890412233
optimizer_0/__6_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bb970a962acd0c2436f95809660e425cabb0e19002a6e51d57380f762846ff4
3
+ size 6890303892
optimizer_0/__7_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afa4bd02ca94b6e6a6482a85237fb17aa7dfd0336de4a1c9f767a1451203d0bd
3
+ size 6890303892
pytorch_model_fsdp_0/.metadata ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85559745b552c64f360dfc2d14bd0896695bbf9bbd4db40cbdd0c9b2fb3b9f63
3
+ size 71442887
pytorch_model_fsdp_0/__0_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfba0ddcc31fbfe70ce8326f47114be646a7cc08f230afba400481673de02d8e
3
+ size 3440942304
pytorch_model_fsdp_0/__1_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f3fe1bb7d27131891245863227deba001df09e63f1cef8693b89aeee3d07287
3
+ size 3440942304
pytorch_model_fsdp_0/__2_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b050ba4793fcbab58f1d6abd2ed4e47a61f13d337cb29e24705cbac6c6c1ea4
3
+ size 3440942304
pytorch_model_fsdp_0/__3_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25d3ad11225d4d1fe44ed12359940fbaccef87290c6ac574567041572481ab10
3
+ size 3440942304
pytorch_model_fsdp_0/__4_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12d731f048d1ca1b2a3419c2b0b119fa134508a9df747890c51e4229788406d6
3
+ size 3440942304
pytorch_model_fsdp_0/__5_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81ed4d98b620cb83968471c4503b6afd4d83c05178e9e92d00a1c89eed3479e0
3
+ size 3440942304
pytorch_model_fsdp_0/__6_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af3ee82e72249ed2c933cea723e70c9ad953bc930d8834a4e32d05349644f868
3
+ size 3440942304
pytorch_model_fsdp_0/__7_0.distcp ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52b4ff69148418765396f43dbb906054b738e32265321877a1059481396a782e
3
+ size 3440942304
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:217874e62924dcd80c7860f91ad25833d586b956509df365ce061b8db0f9160c
3
+ size 16389
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:691c27f87b90d8f675c52c24ba83540a5e5e1182b9be77c726f725711fe06d64
3
+ size 16389
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e614863fc968f81881363baca60fd6ea92ccacb9ea69096e05b7f42aedd583a
3
+ size 16389
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:003bb43c4a2306ebd012a8c1c785a8953a65f36099e021af2490934576b6ce07
3
+ size 16389
rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8f1529f4da2b5741b5f4f13dc56ed126839c1cb214732f5cd4e429dd832926e
3
+ size 16389
rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c2a7952055f29ea1df0be4451284f6ec3b9bc79e4dca4f0dbd085a08cbc7cc2
3
+ size 16389
rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:906644e47c823eb3053261ba29c2657cc57b44be475dd9eaace58aba075a6575
3
+ size 16389
rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96edc3c86d1bb2b9caca969f8fd8ec25810c96bcd0eb262c28c3cb2d550386a6
3
+ size 16389
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc9a5ad86de34f68a1327b48572afadc77d321d101d7108ac54e5a0cd4e85499
3
+ size 1529
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
tokenizer_config.json ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
231
+ "clean_up_tokenization_spaces": false,
232
+ "eos_token": "<|im_end|>",
233
+ "errors": "replace",
234
+ "extra_special_tokens": {},
235
+ "model_max_length": 131072,
236
+ "pad_token": "<|endoftext|>",
237
+ "split_special_tokens": false,
238
+ "tokenizer_class": "Qwen2Tokenizer",
239
+ "unk_token": null
240
+ }
trainer_state.json ADDED
@@ -0,0 +1,2764 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0025706940874035,
6
+ "eval_steps": 500,
7
+ "global_step": 390,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.002570694087403599,
14
+ "grad_norm": 0.08436182141304016,
15
+ "learning_rate": 2.6315789473684213e-07,
16
+ "loss": 0.9321,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.005141388174807198,
21
+ "grad_norm": 0.08981559425592422,
22
+ "learning_rate": 5.263157894736843e-07,
23
+ "loss": 0.9834,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.007712082262210797,
28
+ "grad_norm": 0.10427704453468323,
29
+ "learning_rate": 7.894736842105264e-07,
30
+ "loss": 1.0591,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.010282776349614395,
35
+ "grad_norm": 0.09508734196424484,
36
+ "learning_rate": 1.0526315789473685e-06,
37
+ "loss": 0.8936,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.012853470437017995,
42
+ "grad_norm": 0.09615321457386017,
43
+ "learning_rate": 1.3157894736842106e-06,
44
+ "loss": 1.1475,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.015424164524421594,
49
+ "grad_norm": 0.08302412927150726,
50
+ "learning_rate": 1.5789473684210528e-06,
51
+ "loss": 0.9795,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.017994858611825194,
56
+ "grad_norm": 0.09016848355531693,
57
+ "learning_rate": 1.842105263157895e-06,
58
+ "loss": 0.8286,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.02056555269922879,
63
+ "grad_norm": 0.10795813798904419,
64
+ "learning_rate": 2.105263157894737e-06,
65
+ "loss": 1.0537,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.02313624678663239,
70
+ "grad_norm": 0.09489724785089493,
71
+ "learning_rate": 2.3684210526315793e-06,
72
+ "loss": 1.0142,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.02570694087403599,
77
+ "grad_norm": 0.08715584129095078,
78
+ "learning_rate": 2.631578947368421e-06,
79
+ "loss": 0.9648,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.028277634961439587,
84
+ "grad_norm": 0.08880320936441422,
85
+ "learning_rate": 2.8947368421052634e-06,
86
+ "loss": 0.9702,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.030848329048843187,
91
+ "grad_norm": 0.13510990142822266,
92
+ "learning_rate": 3.1578947368421056e-06,
93
+ "loss": 0.9951,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.033419023136246784,
98
+ "grad_norm": 0.09561849385499954,
99
+ "learning_rate": 3.421052631578948e-06,
100
+ "loss": 0.9883,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.03598971722365039,
105
+ "grad_norm": 0.10157331079244614,
106
+ "learning_rate": 3.68421052631579e-06,
107
+ "loss": 1.0356,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.038560411311053984,
112
+ "grad_norm": 0.07794145494699478,
113
+ "learning_rate": 3.947368421052632e-06,
114
+ "loss": 0.957,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.04113110539845758,
119
+ "grad_norm": 0.07956437766551971,
120
+ "learning_rate": 4.210526315789474e-06,
121
+ "loss": 1.0615,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.043701799485861184,
126
+ "grad_norm": 0.07735664397478104,
127
+ "learning_rate": 4.473684210526316e-06,
128
+ "loss": 0.9443,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.04627249357326478,
133
+ "grad_norm": 0.07957728952169418,
134
+ "learning_rate": 4.736842105263159e-06,
135
+ "loss": 1.043,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.04884318766066838,
140
+ "grad_norm": 0.06874791532754898,
141
+ "learning_rate": 5e-06,
142
+ "loss": 0.9434,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.05141388174807198,
147
+ "grad_norm": 0.07967688143253326,
148
+ "learning_rate": 5.263157894736842e-06,
149
+ "loss": 1.0703,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.05398457583547558,
154
+ "grad_norm": 0.06457515805959702,
155
+ "learning_rate": 5.526315789473684e-06,
156
+ "loss": 1.0767,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.056555269922879174,
161
+ "grad_norm": 0.062354326248168945,
162
+ "learning_rate": 5.789473684210527e-06,
163
+ "loss": 0.96,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.05912596401028278,
168
+ "grad_norm": 0.057665012776851654,
169
+ "learning_rate": 6.0526315789473685e-06,
170
+ "loss": 1.0117,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.061696658097686374,
175
+ "grad_norm": 0.06438912451267242,
176
+ "learning_rate": 6.315789473684211e-06,
177
+ "loss": 0.958,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.06426735218508997,
182
+ "grad_norm": 0.06496331095695496,
183
+ "learning_rate": 6.578947368421053e-06,
184
+ "loss": 1.0142,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.06683804627249357,
189
+ "grad_norm": 0.061105743050575256,
190
+ "learning_rate": 6.842105263157896e-06,
191
+ "loss": 0.9189,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.06940874035989718,
196
+ "grad_norm": 0.05827053263783455,
197
+ "learning_rate": 7.105263157894737e-06,
198
+ "loss": 0.9414,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.07197943444730077,
203
+ "grad_norm": 0.07196812331676483,
204
+ "learning_rate": 7.36842105263158e-06,
205
+ "loss": 0.9814,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.07455012853470437,
210
+ "grad_norm": 0.06901415437459946,
211
+ "learning_rate": 7.631578947368421e-06,
212
+ "loss": 0.9839,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.07712082262210797,
217
+ "grad_norm": 0.06619318574666977,
218
+ "learning_rate": 7.894736842105265e-06,
219
+ "loss": 0.9443,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.07969151670951156,
224
+ "grad_norm": 0.06467492878437042,
225
+ "learning_rate": 8.157894736842105e-06,
226
+ "loss": 0.9785,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.08226221079691516,
231
+ "grad_norm": 0.06422286480665207,
232
+ "learning_rate": 8.421052631578948e-06,
233
+ "loss": 1.043,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.08483290488431877,
238
+ "grad_norm": 0.05858265608549118,
239
+ "learning_rate": 8.68421052631579e-06,
240
+ "loss": 0.9551,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.08740359897172237,
245
+ "grad_norm": 0.05510206148028374,
246
+ "learning_rate": 8.947368421052632e-06,
247
+ "loss": 0.8755,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.08997429305912596,
252
+ "grad_norm": 0.060953687876462936,
253
+ "learning_rate": 9.210526315789475e-06,
254
+ "loss": 0.8745,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.09254498714652956,
259
+ "grad_norm": 0.06641632318496704,
260
+ "learning_rate": 9.473684210526317e-06,
261
+ "loss": 0.8706,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.09511568123393316,
266
+ "grad_norm": 0.058570634573698044,
267
+ "learning_rate": 9.736842105263159e-06,
268
+ "loss": 0.9771,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.09768637532133675,
273
+ "grad_norm": 0.06269422173500061,
274
+ "learning_rate": 1e-05,
275
+ "loss": 0.9067,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.10025706940874037,
280
+ "grad_norm": 0.11592331528663635,
281
+ "learning_rate": 9.998782302800705e-06,
282
+ "loss": 0.9648,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.10282776349614396,
287
+ "grad_norm": 0.05759728327393532,
288
+ "learning_rate": 9.997561636412896e-06,
289
+ "loss": 0.9253,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.10539845758354756,
294
+ "grad_norm": 0.05665018409490585,
295
+ "learning_rate": 9.996337989963381e-06,
296
+ "loss": 0.8413,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.10796915167095116,
301
+ "grad_norm": 0.06908894330263138,
302
+ "learning_rate": 9.995111352525802e-06,
303
+ "loss": 0.9404,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.11053984575835475,
308
+ "grad_norm": 0.05868048593401909,
309
+ "learning_rate": 9.993881713120325e-06,
310
+ "loss": 0.9863,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.11311053984575835,
315
+ "grad_norm": 0.054700467735528946,
316
+ "learning_rate": 9.992649060713315e-06,
317
+ "loss": 0.8315,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.11568123393316196,
322
+ "grad_norm": 0.05400869995355606,
323
+ "learning_rate": 9.991413384216984e-06,
324
+ "loss": 0.9678,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.11825192802056556,
329
+ "grad_norm": 0.097419373691082,
330
+ "learning_rate": 9.990174672489083e-06,
331
+ "loss": 0.9155,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.12082262210796915,
336
+ "grad_norm": 0.056131381541490555,
337
+ "learning_rate": 9.98893291433256e-06,
338
+ "loss": 0.8457,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.12339331619537275,
343
+ "grad_norm": 0.05909228324890137,
344
+ "learning_rate": 9.987688098495213e-06,
345
+ "loss": 1.0063,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.12596401028277635,
350
+ "grad_norm": 0.05639795586466789,
351
+ "learning_rate": 9.986440213669361e-06,
352
+ "loss": 0.9312,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.12853470437017994,
357
+ "grad_norm": 0.05743168666958809,
358
+ "learning_rate": 9.985189248491498e-06,
359
+ "loss": 1.0137,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.13110539845758354,
364
+ "grad_norm": 0.05487934499979019,
365
+ "learning_rate": 9.983935191541947e-06,
366
+ "loss": 0.9834,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.13367609254498714,
371
+ "grad_norm": 0.06367425620555878,
372
+ "learning_rate": 9.982678031344515e-06,
373
+ "loss": 0.9819,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.13624678663239073,
378
+ "grad_norm": 0.05434500426054001,
379
+ "learning_rate": 9.98141775636614e-06,
380
+ "loss": 0.981,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.13881748071979436,
385
+ "grad_norm": 0.0519779697060585,
386
+ "learning_rate": 9.980154355016539e-06,
387
+ "loss": 0.853,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.14138817480719795,
392
+ "grad_norm": 0.053186558187007904,
393
+ "learning_rate": 9.978887815647856e-06,
394
+ "loss": 1.043,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.14395886889460155,
399
+ "grad_norm": 0.06038914993405342,
400
+ "learning_rate": 9.977618126554297e-06,
401
+ "loss": 1.0234,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.14652956298200515,
406
+ "grad_norm": 0.061486050486564636,
407
+ "learning_rate": 9.976345275971782e-06,
408
+ "loss": 0.999,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.14910025706940874,
413
+ "grad_norm": 0.05314328148961067,
414
+ "learning_rate": 9.975069252077563e-06,
415
+ "loss": 0.8911,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.15167095115681234,
420
+ "grad_norm": 0.05191374570131302,
421
+ "learning_rate": 9.973790042989877e-06,
422
+ "loss": 0.9053,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.15424164524421594,
427
+ "grad_norm": 0.0591641366481781,
428
+ "learning_rate": 9.972507636767566e-06,
429
+ "loss": 0.8862,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.15681233933161953,
434
+ "grad_norm": 0.055462323129177094,
435
+ "learning_rate": 9.971222021409705e-06,
436
+ "loss": 0.8442,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.15938303341902313,
441
+ "grad_norm": 0.05295965448021889,
442
+ "learning_rate": 9.969933184855235e-06,
443
+ "loss": 0.8765,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.16195372750642673,
448
+ "grad_norm": 0.08050302416086197,
449
+ "learning_rate": 9.96864111498258e-06,
450
+ "loss": 0.8735,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.16452442159383032,
455
+ "grad_norm": 0.07282618433237076,
456
+ "learning_rate": 9.967345799609267e-06,
457
+ "loss": 0.9307,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.16709511568123395,
462
+ "grad_norm": 0.05436393991112709,
463
+ "learning_rate": 9.966047226491548e-06,
464
+ "loss": 0.8174,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.16966580976863754,
469
+ "grad_norm": 0.05547136813402176,
470
+ "learning_rate": 9.964745383324008e-06,
471
+ "loss": 0.8633,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.17223650385604114,
476
+ "grad_norm": 0.05089852213859558,
477
+ "learning_rate": 9.96344025773918e-06,
478
+ "loss": 0.8999,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.17480719794344474,
483
+ "grad_norm": 0.06282460689544678,
484
+ "learning_rate": 9.962131837307155e-06,
485
+ "loss": 0.8545,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.17737789203084833,
490
+ "grad_norm": 0.054014649242162704,
491
+ "learning_rate": 9.960820109535178e-06,
492
+ "loss": 1.0195,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.17994858611825193,
497
+ "grad_norm": 0.05466762185096741,
498
+ "learning_rate": 9.959505061867267e-06,
499
+ "loss": 0.9785,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.18251928020565553,
504
+ "grad_norm": 0.05074343457818031,
505
+ "learning_rate": 9.958186681683795e-06,
506
+ "loss": 0.853,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.18508997429305912,
511
+ "grad_norm": 0.055692434310913086,
512
+ "learning_rate": 9.9568649563011e-06,
513
+ "loss": 0.8608,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.18766066838046272,
518
+ "grad_norm": 0.053501665592193604,
519
+ "learning_rate": 9.955539872971066e-06,
520
+ "loss": 1.083,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.19023136246786632,
525
+ "grad_norm": 0.0508846677839756,
526
+ "learning_rate": 9.954211418880725e-06,
527
+ "loss": 0.9199,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.1928020565552699,
532
+ "grad_norm": 0.049147725105285645,
533
+ "learning_rate": 9.952879581151834e-06,
534
+ "loss": 0.9028,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.1953727506426735,
539
+ "grad_norm": 0.05541389435529709,
540
+ "learning_rate": 9.951544346840466e-06,
541
+ "loss": 0.9097,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.19794344473007713,
546
+ "grad_norm": 0.05445914342999458,
547
+ "learning_rate": 9.950205702936587e-06,
548
+ "loss": 1.0361,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.20051413881748073,
553
+ "grad_norm": 0.051683664321899414,
554
+ "learning_rate": 9.948863636363638e-06,
555
+ "loss": 0.8433,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.20308483290488433,
560
+ "grad_norm": 0.05170630291104317,
561
+ "learning_rate": 9.947518133978098e-06,
562
+ "loss": 0.9438,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.20565552699228792,
567
+ "grad_norm": 0.057460807263851166,
568
+ "learning_rate": 9.94616918256907e-06,
569
+ "loss": 0.8032,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.20822622107969152,
574
+ "grad_norm": 0.1279779076576233,
575
+ "learning_rate": 9.944816768857836e-06,
576
+ "loss": 1.019,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.21079691516709512,
581
+ "grad_norm": 0.048682596534490585,
582
+ "learning_rate": 9.943460879497431e-06,
583
+ "loss": 0.9692,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.2133676092544987,
588
+ "grad_norm": 0.04913777485489845,
589
+ "learning_rate": 9.942101501072196e-06,
590
+ "loss": 0.8569,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.2159383033419023,
595
+ "grad_norm": 0.054994840174913406,
596
+ "learning_rate": 9.940738620097337e-06,
597
+ "loss": 0.876,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.2185089974293059,
602
+ "grad_norm": 0.05107712373137474,
603
+ "learning_rate": 9.939372223018491e-06,
604
+ "loss": 0.9136,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.2210796915167095,
609
+ "grad_norm": 0.05990511178970337,
610
+ "learning_rate": 9.938002296211251e-06,
611
+ "loss": 0.9683,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.2236503856041131,
616
+ "grad_norm": 0.05146004632115364,
617
+ "learning_rate": 9.936628825980745e-06,
618
+ "loss": 0.8701,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.2262210796915167,
623
+ "grad_norm": 0.051048219203948975,
624
+ "learning_rate": 9.93525179856115e-06,
625
+ "loss": 0.896,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.22879177377892032,
630
+ "grad_norm": 0.054575320333242416,
631
+ "learning_rate": 9.933871200115258e-06,
632
+ "loss": 0.8853,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.23136246786632392,
637
+ "grad_norm": 0.0543404147028923,
638
+ "learning_rate": 9.932487016733988e-06,
639
+ "loss": 0.9272,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.23393316195372751,
644
+ "grad_norm": 0.08317338675260544,
645
+ "learning_rate": 9.931099234435939e-06,
646
+ "loss": 0.8892,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.2365038560411311,
651
+ "grad_norm": 0.051799483597278595,
652
+ "learning_rate": 9.929707839166908e-06,
653
+ "loss": 0.9038,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.2390745501285347,
658
+ "grad_norm": 0.0485287643969059,
659
+ "learning_rate": 9.928312816799421e-06,
660
+ "loss": 0.8154,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.2416452442159383,
665
+ "grad_norm": 0.04724659398198128,
666
+ "learning_rate": 9.92691415313225e-06,
667
+ "loss": 0.7871,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.2442159383033419,
672
+ "grad_norm": 0.051631320267915726,
673
+ "learning_rate": 9.925511833889939e-06,
674
+ "loss": 0.9146,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.2467866323907455,
679
+ "grad_norm": 0.05385703966021538,
680
+ "learning_rate": 9.924105844722303e-06,
681
+ "loss": 0.9297,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.2493573264781491,
686
+ "grad_norm": 0.09320125728845596,
687
+ "learning_rate": 9.922696171203961e-06,
688
+ "loss": 0.8818,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.2519280205655527,
693
+ "grad_norm": 0.05302536487579346,
694
+ "learning_rate": 9.921282798833818e-06,
695
+ "loss": 0.8369,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.2544987146529563,
700
+ "grad_norm": 0.05064469203352928,
701
+ "learning_rate": 9.919865713034594e-06,
702
+ "loss": 0.9038,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.2570694087403599,
707
+ "grad_norm": 0.055192429572343826,
708
+ "learning_rate": 9.918444899152296e-06,
709
+ "loss": 0.8843,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.2596401028277635,
714
+ "grad_norm": 0.0517449676990509,
715
+ "learning_rate": 9.91702034245573e-06,
716
+ "loss": 0.9277,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.2622107969151671,
721
+ "grad_norm": 0.062362704426050186,
722
+ "learning_rate": 9.915592028135992e-06,
723
+ "loss": 0.8643,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.2647814910025707,
728
+ "grad_norm": 0.05347563698887825,
729
+ "learning_rate": 9.914159941305944e-06,
730
+ "loss": 0.8767,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.26735218508997427,
735
+ "grad_norm": 0.05177872255444527,
736
+ "learning_rate": 9.912724066999706e-06,
737
+ "loss": 0.8186,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.2699228791773779,
742
+ "grad_norm": 0.05178435519337654,
743
+ "learning_rate": 9.911284390172136e-06,
744
+ "loss": 0.8301,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.27249357326478146,
749
+ "grad_norm": 0.05058378726243973,
750
+ "learning_rate": 9.909840895698292e-06,
751
+ "loss": 0.8242,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.2750642673521851,
756
+ "grad_norm": 0.05525752529501915,
757
+ "learning_rate": 9.908393568372917e-06,
758
+ "loss": 0.9287,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.2776349614395887,
763
+ "grad_norm": 0.056954048573970795,
764
+ "learning_rate": 9.906942392909897e-06,
765
+ "loss": 0.8618,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.2802056555269923,
770
+ "grad_norm": 0.052674438804388046,
771
+ "learning_rate": 9.905487353941726e-06,
772
+ "loss": 0.8745,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.2827763496143959,
777
+ "grad_norm": 0.05273294448852539,
778
+ "learning_rate": 9.904028436018958e-06,
779
+ "loss": 0.9653,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.2853470437017995,
784
+ "grad_norm": 0.05354757234454155,
785
+ "learning_rate": 9.90256562360967e-06,
786
+ "loss": 0.8643,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.2879177377892031,
791
+ "grad_norm": 0.06856588274240494,
792
+ "learning_rate": 9.901098901098903e-06,
793
+ "loss": 0.8179,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.29048843187660667,
798
+ "grad_norm": 0.05660005658864975,
799
+ "learning_rate": 9.899628252788105e-06,
800
+ "loss": 0.8325,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.2930591259640103,
805
+ "grad_norm": 0.05592386797070503,
806
+ "learning_rate": 9.89815366289458e-06,
807
+ "loss": 0.9209,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.29562982005141386,
812
+ "grad_norm": 0.05370522290468216,
813
+ "learning_rate": 9.896675115550918e-06,
814
+ "loss": 0.9277,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.2982005141388175,
819
+ "grad_norm": 0.05046721547842026,
820
+ "learning_rate": 9.895192594804421e-06,
821
+ "loss": 0.9165,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.30077120822622105,
826
+ "grad_norm": 0.04982549697160721,
827
+ "learning_rate": 9.893706084616536e-06,
828
+ "loss": 0.7983,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.3033419023136247,
833
+ "grad_norm": 0.06228173151612282,
834
+ "learning_rate": 9.892215568862277e-06,
835
+ "loss": 0.9219,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.3059125964010283,
840
+ "grad_norm": 0.055755749344825745,
841
+ "learning_rate": 9.890721031329637e-06,
842
+ "loss": 0.8994,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.30848329048843187,
847
+ "grad_norm": 0.05125127360224724,
848
+ "learning_rate": 9.889222455719005e-06,
849
+ "loss": 0.8315,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.3110539845758355,
854
+ "grad_norm": 0.060036104172468185,
855
+ "learning_rate": 9.887719825642569e-06,
856
+ "loss": 0.9575,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.31362467866323906,
861
+ "grad_norm": 0.05109132453799248,
862
+ "learning_rate": 9.886213124623721e-06,
863
+ "loss": 0.9683,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.3161953727506427,
868
+ "grad_norm": 0.050311293452978134,
869
+ "learning_rate": 9.884702336096457e-06,
870
+ "loss": 1.002,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.31876606683804626,
875
+ "grad_norm": 0.051465220749378204,
876
+ "learning_rate": 9.88318744340477e-06,
877
+ "loss": 0.959,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.3213367609254499,
882
+ "grad_norm": 0.055082421749830246,
883
+ "learning_rate": 9.881668429802025e-06,
884
+ "loss": 0.9424,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.32390745501285345,
889
+ "grad_norm": 0.05376673862338066,
890
+ "learning_rate": 9.880145278450364e-06,
891
+ "loss": 0.833,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.3264781491002571,
896
+ "grad_norm": 0.05093827843666077,
897
+ "learning_rate": 9.878617972420063e-06,
898
+ "loss": 0.9312,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.32904884318766064,
903
+ "grad_norm": 0.045088816434144974,
904
+ "learning_rate": 9.877086494688923e-06,
905
+ "loss": 0.8423,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.33161953727506427,
910
+ "grad_norm": 0.049116652458906174,
911
+ "learning_rate": 9.87555082814162e-06,
912
+ "loss": 0.8467,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.3341902313624679,
917
+ "grad_norm": 0.052838873118162155,
918
+ "learning_rate": 9.87401095556908e-06,
919
+ "loss": 0.9038,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.33676092544987146,
924
+ "grad_norm": 0.051667746156454086,
925
+ "learning_rate": 9.872466859667836e-06,
926
+ "loss": 0.8433,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.3393316195372751,
931
+ "grad_norm": 0.051519960165023804,
932
+ "learning_rate": 9.870918523039367e-06,
933
+ "loss": 0.8535,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.34190231362467866,
938
+ "grad_norm": 0.05033135414123535,
939
+ "learning_rate": 9.869365928189458e-06,
940
+ "loss": 1.0623,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.3444730077120823,
945
+ "grad_norm": 0.05721587687730789,
946
+ "learning_rate": 9.86780905752754e-06,
947
+ "loss": 0.8481,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.34704370179948585,
952
+ "grad_norm": 0.05643352121114731,
953
+ "learning_rate": 9.86624789336602e-06,
954
+ "loss": 0.9263,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.3496143958868895,
959
+ "grad_norm": 0.051524702459573746,
960
+ "learning_rate": 9.864682417919608e-06,
961
+ "loss": 0.8252,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.35218508997429304,
966
+ "grad_norm": 0.050419341772794724,
967
+ "learning_rate": 9.863112613304656e-06,
968
+ "loss": 0.8091,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.35475578406169667,
973
+ "grad_norm": 0.04712434485554695,
974
+ "learning_rate": 9.861538461538462e-06,
975
+ "loss": 0.9023,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.35732647814910024,
980
+ "grad_norm": 0.05018655210733414,
981
+ "learning_rate": 9.859959944538593e-06,
982
+ "loss": 0.8362,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.35989717223650386,
987
+ "grad_norm": 0.0635514110326767,
988
+ "learning_rate": 9.858377044122186e-06,
989
+ "loss": 0.8818,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.36246786632390743,
994
+ "grad_norm": 0.06259825825691223,
995
+ "learning_rate": 9.856789742005253e-06,
996
+ "loss": 0.8423,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.36503856041131105,
1001
+ "grad_norm": 0.05616484954953194,
1002
+ "learning_rate": 9.85519801980198e-06,
1003
+ "loss": 0.8696,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.3676092544987147,
1008
+ "grad_norm": 0.05006776750087738,
1009
+ "learning_rate": 9.853601859024013e-06,
1010
+ "loss": 0.9873,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.37017994858611825,
1015
+ "grad_norm": 0.04944387823343277,
1016
+ "learning_rate": 9.85200124107974e-06,
1017
+ "loss": 0.9453,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.37275064267352187,
1022
+ "grad_norm": 0.048190075904130936,
1023
+ "learning_rate": 9.850396147273575e-06,
1024
+ "loss": 0.8677,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.37532133676092544,
1029
+ "grad_norm": 0.1091238260269165,
1030
+ "learning_rate": 9.848786558805228e-06,
1031
+ "loss": 0.8442,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.37789203084832906,
1036
+ "grad_norm": 0.04897330328822136,
1037
+ "learning_rate": 9.847172456768967e-06,
1038
+ "loss": 0.9849,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.38046272493573263,
1043
+ "grad_norm": 0.05173799395561218,
1044
+ "learning_rate": 9.845553822152887e-06,
1045
+ "loss": 0.8701,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.38303341902313626,
1050
+ "grad_norm": 0.058186061680316925,
1051
+ "learning_rate": 9.843930635838151e-06,
1052
+ "loss": 0.9551,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.3856041131105398,
1057
+ "grad_norm": 0.06420359015464783,
1058
+ "learning_rate": 9.84230287859825e-06,
1059
+ "loss": 0.9351,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.38817480719794345,
1064
+ "grad_norm": 0.05216103047132492,
1065
+ "learning_rate": 9.840670531098232e-06,
1066
+ "loss": 0.9629,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.390745501285347,
1071
+ "grad_norm": 0.06294068694114685,
1072
+ "learning_rate": 9.839033573893945e-06,
1073
+ "loss": 0.8647,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.39331619537275064,
1078
+ "grad_norm": 0.05219503119587898,
1079
+ "learning_rate": 9.837391987431266e-06,
1080
+ "loss": 0.8042,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.39588688946015427,
1085
+ "grad_norm": 0.052275657653808594,
1086
+ "learning_rate": 9.835745752045313e-06,
1087
+ "loss": 0.9263,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.39845758354755784,
1092
+ "grad_norm": 0.04824310168623924,
1093
+ "learning_rate": 9.834094847959667e-06,
1094
+ "loss": 0.9258,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.40102827763496146,
1099
+ "grad_norm": 0.05650007724761963,
1100
+ "learning_rate": 9.83243925528558e-06,
1101
+ "loss": 0.957,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.40359897172236503,
1106
+ "grad_norm": 0.09102229028940201,
1107
+ "learning_rate": 9.830778954021173e-06,
1108
+ "loss": 0.9434,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.40616966580976865,
1113
+ "grad_norm": 0.05498167499899864,
1114
+ "learning_rate": 9.829113924050634e-06,
1115
+ "loss": 0.9219,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.4087403598971722,
1120
+ "grad_norm": 0.053989022970199585,
1121
+ "learning_rate": 9.8274441451434e-06,
1122
+ "loss": 0.9297,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.41131105398457585,
1127
+ "grad_norm": 0.055620986968278885,
1128
+ "learning_rate": 9.82576959695335e-06,
1129
+ "loss": 0.9141,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.4138817480719794,
1134
+ "grad_norm": 0.05333478003740311,
1135
+ "learning_rate": 9.824090259017958e-06,
1136
+ "loss": 0.9053,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.41645244215938304,
1141
+ "grad_norm": 0.048174161463975906,
1142
+ "learning_rate": 9.82240611075748e-06,
1143
+ "loss": 0.7944,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.4190231362467866,
1148
+ "grad_norm": 0.05324098467826843,
1149
+ "learning_rate": 9.820717131474105e-06,
1150
+ "loss": 0.8779,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.42159383033419023,
1155
+ "grad_norm": 0.05260465666651726,
1156
+ "learning_rate": 9.819023300351102e-06,
1157
+ "loss": 1.0474,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.4241645244215938,
1162
+ "grad_norm": 0.04889111965894699,
1163
+ "learning_rate": 9.817324596451974e-06,
1164
+ "loss": 0.8452,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.4267352185089974,
1169
+ "grad_norm": 0.052814241498708725,
1170
+ "learning_rate": 9.81562099871959e-06,
1171
+ "loss": 0.9253,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.42930591259640105,
1176
+ "grad_norm": 0.05090152099728584,
1177
+ "learning_rate": 9.813912485975318e-06,
1178
+ "loss": 0.8994,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.4318766066838046,
1183
+ "grad_norm": 0.05313080549240112,
1184
+ "learning_rate": 9.812199036918139e-06,
1185
+ "loss": 0.9429,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.43444730077120824,
1190
+ "grad_norm": 0.059309348464012146,
1191
+ "learning_rate": 9.810480630123775e-06,
1192
+ "loss": 0.9697,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.4370179948586118,
1197
+ "grad_norm": 0.05176108330488205,
1198
+ "learning_rate": 9.808757244043786e-06,
1199
+ "loss": 0.9204,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.43958868894601544,
1204
+ "grad_norm": 0.05403340235352516,
1205
+ "learning_rate": 9.807028857004676e-06,
1206
+ "loss": 0.9077,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.442159383033419,
1211
+ "grad_norm": 0.05564318597316742,
1212
+ "learning_rate": 9.805295447206974e-06,
1213
+ "loss": 0.9058,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.44473007712082263,
1218
+ "grad_norm": 0.049466319382190704,
1219
+ "learning_rate": 9.803556992724334e-06,
1220
+ "loss": 0.8521,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.4473007712082262,
1225
+ "grad_norm": 0.05796355754137039,
1226
+ "learning_rate": 9.80181347150259e-06,
1227
+ "loss": 0.9028,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.4498714652956298,
1232
+ "grad_norm": 0.052372533828020096,
1233
+ "learning_rate": 9.800064861358847e-06,
1234
+ "loss": 0.9624,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.4524421593830334,
1239
+ "grad_norm": 0.04954168573021889,
1240
+ "learning_rate": 9.798311139980513e-06,
1241
+ "loss": 0.8271,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.455012853470437,
1246
+ "grad_norm": 0.051455263048410416,
1247
+ "learning_rate": 9.79655228492438e-06,
1248
+ "loss": 0.814,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.45758354755784064,
1253
+ "grad_norm": 0.0496753565967083,
1254
+ "learning_rate": 9.794788273615637e-06,
1255
+ "loss": 0.7856,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.4601542416452442,
1260
+ "grad_norm": 0.05122137442231178,
1261
+ "learning_rate": 9.793019083346927e-06,
1262
+ "loss": 0.7886,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.46272493573264784,
1267
+ "grad_norm": 0.05189337953925133,
1268
+ "learning_rate": 9.79124469127736e-06,
1269
+ "loss": 0.8931,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.4652956298200514,
1274
+ "grad_norm": 0.052611690014600754,
1275
+ "learning_rate": 9.78946507443154e-06,
1276
+ "loss": 0.8657,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.46786632390745503,
1281
+ "grad_norm": 0.05381154641509056,
1282
+ "learning_rate": 9.787680209698559e-06,
1283
+ "loss": 0.9116,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.4704370179948586,
1288
+ "grad_norm": 0.05444910377264023,
1289
+ "learning_rate": 9.78589007383101e-06,
1290
+ "loss": 0.8716,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.4730077120822622,
1295
+ "grad_norm": 0.05494491383433342,
1296
+ "learning_rate": 9.78409464344397e-06,
1297
+ "loss": 0.8936,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.4755784061696658,
1302
+ "grad_norm": 0.052697766572237015,
1303
+ "learning_rate": 9.78229389501399e-06,
1304
+ "loss": 0.9004,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.4781491002570694,
1309
+ "grad_norm": 0.05403268709778786,
1310
+ "learning_rate": 9.780487804878049e-06,
1311
+ "loss": 0.9561,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.480719794344473,
1316
+ "grad_norm": 0.05628068372607231,
1317
+ "learning_rate": 9.778676349232548e-06,
1318
+ "loss": 0.8818,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.4832904884318766,
1323
+ "grad_norm": 0.04996819794178009,
1324
+ "learning_rate": 9.776859504132232e-06,
1325
+ "loss": 0.8843,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.48586118251928023,
1330
+ "grad_norm": 0.06892653554677963,
1331
+ "learning_rate": 9.775037245489159e-06,
1332
+ "loss": 0.9004,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.4884318766066838,
1337
+ "grad_norm": 0.12632861733436584,
1338
+ "learning_rate": 9.773209549071618e-06,
1339
+ "loss": 1.0562,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.4910025706940874,
1344
+ "grad_norm": 0.05255817621946335,
1345
+ "learning_rate": 9.771376390503072e-06,
1346
+ "loss": 0.9155,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.493573264781491,
1351
+ "grad_norm": 0.07260231673717499,
1352
+ "learning_rate": 9.76953774526106e-06,
1353
+ "loss": 0.8818,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.4961439588688946,
1358
+ "grad_norm": 0.053487952798604965,
1359
+ "learning_rate": 9.767693588676105e-06,
1360
+ "loss": 0.938,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.4987146529562982,
1365
+ "grad_norm": 0.05333535373210907,
1366
+ "learning_rate": 9.76584389593062e-06,
1367
+ "loss": 0.8901,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.5012853470437018,
1372
+ "grad_norm": 0.052906159311532974,
1373
+ "learning_rate": 9.763988642057791e-06,
1374
+ "loss": 0.9014,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.5038560411311054,
1379
+ "grad_norm": 0.06862504035234451,
1380
+ "learning_rate": 9.762127801940448e-06,
1381
+ "loss": 0.7451,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.506426735218509,
1386
+ "grad_norm": 0.05256706848740578,
1387
+ "learning_rate": 9.760261350309935e-06,
1388
+ "loss": 0.8257,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.5089974293059126,
1393
+ "grad_norm": 0.05243777483701706,
1394
+ "learning_rate": 9.758389261744968e-06,
1395
+ "loss": 0.9609,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.5115681233933161,
1400
+ "grad_norm": 0.04881735518574715,
1401
+ "learning_rate": 9.756511510670476e-06,
1402
+ "loss": 0.8149,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.5141388174807198,
1407
+ "grad_norm": 0.051057204604148865,
1408
+ "learning_rate": 9.754628071356445e-06,
1409
+ "loss": 0.8906,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.5167095115681234,
1414
+ "grad_norm": 0.05244697630405426,
1415
+ "learning_rate": 9.752738917916738e-06,
1416
+ "loss": 0.9429,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.519280205655527,
1421
+ "grad_norm": 0.04986201226711273,
1422
+ "learning_rate": 9.750844024307901e-06,
1423
+ "loss": 0.8267,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.5218508997429306,
1428
+ "grad_norm": 0.059997860342264175,
1429
+ "learning_rate": 9.74894336432798e-06,
1430
+ "loss": 0.9497,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.5244215938303342,
1435
+ "grad_norm": 0.05549291893839836,
1436
+ "learning_rate": 9.747036911615308e-06,
1437
+ "loss": 0.9771,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.5269922879177378,
1442
+ "grad_norm": 0.052215415984392166,
1443
+ "learning_rate": 9.745124639647277e-06,
1444
+ "loss": 0.8462,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.5295629820051414,
1449
+ "grad_norm": 0.06002132222056389,
1450
+ "learning_rate": 9.74320652173913e-06,
1451
+ "loss": 0.8823,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.532133676092545,
1456
+ "grad_norm": 0.055552612990140915,
1457
+ "learning_rate": 9.741282531042695e-06,
1458
+ "loss": 0.8638,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.5347043701799485,
1463
+ "grad_norm": 0.058621279895305634,
1464
+ "learning_rate": 9.739352640545145e-06,
1465
+ "loss": 0.9531,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.5372750642673522,
1470
+ "grad_norm": 0.05218435078859329,
1471
+ "learning_rate": 9.737416823067735e-06,
1472
+ "loss": 0.8994,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.5398457583547558,
1477
+ "grad_norm": 0.05027312785387039,
1478
+ "learning_rate": 9.735475051264526e-06,
1479
+ "loss": 0.8926,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.5424164524421594,
1484
+ "grad_norm": 0.05271938070654869,
1485
+ "learning_rate": 9.733527297621086e-06,
1486
+ "loss": 0.96,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.5449871465295629,
1491
+ "grad_norm": 0.061470016837120056,
1492
+ "learning_rate": 9.731573534453205e-06,
1493
+ "loss": 0.9233,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.5475578406169666,
1498
+ "grad_norm": 0.05070161819458008,
1499
+ "learning_rate": 9.72961373390558e-06,
1500
+ "loss": 0.8794,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.5501285347043702,
1505
+ "grad_norm": 0.05650542303919792,
1506
+ "learning_rate": 9.727647867950482e-06,
1507
+ "loss": 0.8892,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.5526992287917738,
1512
+ "grad_norm": 0.05301105976104736,
1513
+ "learning_rate": 9.725675908386431e-06,
1514
+ "loss": 0.8643,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.5552699228791774,
1519
+ "grad_norm": 0.055119194090366364,
1520
+ "learning_rate": 9.723697826836842e-06,
1521
+ "loss": 1.1055,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.5578406169665809,
1526
+ "grad_norm": 0.052684854716062546,
1527
+ "learning_rate": 9.721713594748662e-06,
1528
+ "loss": 0.8765,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.5604113110539846,
1533
+ "grad_norm": 0.060191500931978226,
1534
+ "learning_rate": 9.719723183391004e-06,
1535
+ "loss": 0.9316,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.5629820051413882,
1540
+ "grad_norm": 0.0570666678249836,
1541
+ "learning_rate": 9.71772656385375e-06,
1542
+ "loss": 0.8291,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.5655526992287918,
1547
+ "grad_norm": 0.05079682916402817,
1548
+ "learning_rate": 9.715723707046166e-06,
1549
+ "loss": 0.9233,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.5681233933161953,
1554
+ "grad_norm": 0.05589374899864197,
1555
+ "learning_rate": 9.713714583695464e-06,
1556
+ "loss": 0.9663,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.570694087403599,
1561
+ "grad_norm": 0.049071066081523895,
1562
+ "learning_rate": 9.711699164345404e-06,
1563
+ "loss": 0.9448,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.5732647814910026,
1568
+ "grad_norm": 0.048681292682886124,
1569
+ "learning_rate": 9.709677419354838e-06,
1570
+ "loss": 0.7961,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.5758354755784062,
1575
+ "grad_norm": 0.05470262095332146,
1576
+ "learning_rate": 9.707649318896264e-06,
1577
+ "loss": 0.9468,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.5784061696658098,
1582
+ "grad_norm": 0.05211283266544342,
1583
+ "learning_rate": 9.705614832954347e-06,
1584
+ "loss": 0.8276,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.5809768637532133,
1589
+ "grad_norm": 0.04872957989573479,
1590
+ "learning_rate": 9.703573931324458e-06,
1591
+ "loss": 0.7153,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.583547557840617,
1596
+ "grad_norm": 0.05287624150514603,
1597
+ "learning_rate": 9.70152658361116e-06,
1598
+ "loss": 0.8726,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.5861182519280206,
1603
+ "grad_norm": 0.05211886763572693,
1604
+ "learning_rate": 9.699472759226714e-06,
1605
+ "loss": 0.9302,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.5886889460154242,
1610
+ "grad_norm": 0.05334436520934105,
1611
+ "learning_rate": 9.697412427389544e-06,
1612
+ "loss": 1.0332,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.5912596401028277,
1617
+ "grad_norm": 0.053637877106666565,
1618
+ "learning_rate": 9.695345557122709e-06,
1619
+ "loss": 0.8999,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.5938303341902313,
1624
+ "grad_norm": 0.04681619256734848,
1625
+ "learning_rate": 9.69327211725234e-06,
1626
+ "loss": 0.8076,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.596401028277635,
1631
+ "grad_norm": 0.052718836814165115,
1632
+ "learning_rate": 9.691192076406085e-06,
1633
+ "loss": 0.8906,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.5989717223650386,
1638
+ "grad_norm": 0.05302907153964043,
1639
+ "learning_rate": 9.689105403011516e-06,
1640
+ "loss": 0.937,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.6015424164524421,
1645
+ "grad_norm": 0.05149109661579132,
1646
+ "learning_rate": 9.687012065294536e-06,
1647
+ "loss": 0.8506,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.6041131105398457,
1652
+ "grad_norm": 0.05238654837012291,
1653
+ "learning_rate": 9.68491203127777e-06,
1654
+ "loss": 0.8628,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.6066838046272494,
1659
+ "grad_norm": 0.06081896275281906,
1660
+ "learning_rate": 9.682805268778926e-06,
1661
+ "loss": 0.9048,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.609254498714653,
1666
+ "grad_norm": 0.05550535395741463,
1667
+ "learning_rate": 9.680691745409164e-06,
1668
+ "loss": 0.9307,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.6118251928020566,
1673
+ "grad_norm": 0.05248388275504112,
1674
+ "learning_rate": 9.678571428571429e-06,
1675
+ "loss": 0.9956,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.6143958868894601,
1680
+ "grad_norm": 0.05890145152807236,
1681
+ "learning_rate": 9.676444285458773e-06,
1682
+ "loss": 0.832,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.6169665809768637,
1687
+ "grad_norm": 0.051260773092508316,
1688
+ "learning_rate": 9.67431028305267e-06,
1689
+ "loss": 0.8467,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.6195372750642674,
1694
+ "grad_norm": 0.05052540451288223,
1695
+ "learning_rate": 9.6721693881213e-06,
1696
+ "loss": 0.8301,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.622107969151671,
1701
+ "grad_norm": 0.0522305965423584,
1702
+ "learning_rate": 9.67002156721783e-06,
1703
+ "loss": 0.8354,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.6246786632390745,
1708
+ "grad_norm": 0.0485188327729702,
1709
+ "learning_rate": 9.66786678667867e-06,
1710
+ "loss": 0.8994,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.6272493573264781,
1715
+ "grad_norm": 0.0574263334274292,
1716
+ "learning_rate": 9.66570501262171e-06,
1717
+ "loss": 1.042,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.6298200514138818,
1722
+ "grad_norm": 0.05562338978052139,
1723
+ "learning_rate": 9.663536210944555e-06,
1724
+ "loss": 0.9741,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.6323907455012854,
1729
+ "grad_norm": 0.05564079061150551,
1730
+ "learning_rate": 9.661360347322722e-06,
1731
+ "loss": 0.9346,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.6349614395886889,
1736
+ "grad_norm": 0.053419217467308044,
1737
+ "learning_rate": 9.659177387207828e-06,
1738
+ "loss": 0.9951,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.6375321336760925,
1743
+ "grad_norm": 0.057053521275520325,
1744
+ "learning_rate": 9.656987295825771e-06,
1745
+ "loss": 0.9004,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.6401028277634961,
1750
+ "grad_norm": 0.05263104662299156,
1751
+ "learning_rate": 9.654790038174878e-06,
1752
+ "loss": 0.8872,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 0.6426735218508998,
1757
+ "grad_norm": 0.049765683710575104,
1758
+ "learning_rate": 9.652585579024035e-06,
1759
+ "loss": 0.8008,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 0.6452442159383034,
1764
+ "grad_norm": 0.05088181048631668,
1765
+ "learning_rate": 9.650373882910816e-06,
1766
+ "loss": 0.939,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 0.6478149100257069,
1771
+ "grad_norm": 0.05318666622042656,
1772
+ "learning_rate": 9.648154914139569e-06,
1773
+ "loss": 0.9712,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 0.6503856041131105,
1778
+ "grad_norm": 0.055674273520708084,
1779
+ "learning_rate": 9.645928636779507e-06,
1780
+ "loss": 0.9609,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 0.6529562982005142,
1785
+ "grad_norm": 0.05109839141368866,
1786
+ "learning_rate": 9.643695014662755e-06,
1787
+ "loss": 0.8394,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 0.6555269922879178,
1792
+ "grad_norm": 0.051385875791311264,
1793
+ "learning_rate": 9.641454011382413e-06,
1794
+ "loss": 0.853,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 0.6580976863753213,
1799
+ "grad_norm": 0.05000017583370209,
1800
+ "learning_rate": 9.639205590290549e-06,
1801
+ "loss": 0.8999,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 0.6606683804627249,
1806
+ "grad_norm": 0.0533922053873539,
1807
+ "learning_rate": 9.636949714496225e-06,
1808
+ "loss": 0.9468,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 0.6632390745501285,
1813
+ "grad_norm": 0.057750582695007324,
1814
+ "learning_rate": 9.63468634686347e-06,
1815
+ "loss": 0.8018,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 0.6658097686375322,
1820
+ "grad_norm": 0.05343909561634064,
1821
+ "learning_rate": 9.632415450009242e-06,
1822
+ "loss": 0.9253,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 0.6683804627249358,
1827
+ "grad_norm": 0.15355075895786285,
1828
+ "learning_rate": 9.63013698630137e-06,
1829
+ "loss": 0.8589,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 0.6709511568123393,
1834
+ "grad_norm": 0.05503278225660324,
1835
+ "learning_rate": 9.627850917856482e-06,
1836
+ "loss": 0.9248,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 0.6735218508997429,
1841
+ "grad_norm": 0.055749524384737015,
1842
+ "learning_rate": 9.62555720653789e-06,
1843
+ "loss": 1.0229,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 0.6760925449871465,
1848
+ "grad_norm": 0.05361119285225868,
1849
+ "learning_rate": 9.62325581395349e-06,
1850
+ "loss": 0.876,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 0.6786632390745502,
1855
+ "grad_norm": 0.0522209070622921,
1856
+ "learning_rate": 9.620946701453597e-06,
1857
+ "loss": 0.7866,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 0.6812339331619537,
1862
+ "grad_norm": 0.05726565048098564,
1863
+ "learning_rate": 9.618629830128805e-06,
1864
+ "loss": 0.9082,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 0.6838046272493573,
1869
+ "grad_norm": 0.056867945939302444,
1870
+ "learning_rate": 9.616305160807779e-06,
1871
+ "loss": 1.0034,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 0.6863753213367609,
1876
+ "grad_norm": 0.0533023402094841,
1877
+ "learning_rate": 9.613972654055067e-06,
1878
+ "loss": 0.8677,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 0.6889460154241646,
1883
+ "grad_norm": 0.04977701976895332,
1884
+ "learning_rate": 9.611632270168855e-06,
1885
+ "loss": 0.8765,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 0.6915167095115681,
1890
+ "grad_norm": 0.0571221187710762,
1891
+ "learning_rate": 9.609283969178727e-06,
1892
+ "loss": 0.8521,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 0.6940874035989717,
1897
+ "grad_norm": 0.0479142926633358,
1898
+ "learning_rate": 9.606927710843375e-06,
1899
+ "loss": 1.0391,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 0.6966580976863753,
1904
+ "grad_norm": 0.049262482672929764,
1905
+ "learning_rate": 9.604563454648314e-06,
1906
+ "loss": 0.8076,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 0.699228791773779,
1911
+ "grad_norm": 0.048339374363422394,
1912
+ "learning_rate": 9.602191159803553e-06,
1913
+ "loss": 0.9209,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 0.7017994858611826,
1918
+ "grad_norm": 0.05326506868004799,
1919
+ "learning_rate": 9.59981078524125e-06,
1920
+ "loss": 0.8999,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 0.7043701799485861,
1925
+ "grad_norm": 0.05416352301836014,
1926
+ "learning_rate": 9.597422289613344e-06,
1927
+ "loss": 0.8525,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 0.7069408740359897,
1932
+ "grad_norm": 0.054313864558935165,
1933
+ "learning_rate": 9.595025631289162e-06,
1934
+ "loss": 0.8623,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 0.7095115681233933,
1939
+ "grad_norm": 0.05430547520518303,
1940
+ "learning_rate": 9.592620768352987e-06,
1941
+ "loss": 0.9419,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 0.712082262210797,
1946
+ "grad_norm": 0.05323995649814606,
1947
+ "learning_rate": 9.59020765860164e-06,
1948
+ "loss": 0.7759,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 0.7146529562982005,
1953
+ "grad_norm": 0.05427195131778717,
1954
+ "learning_rate": 9.587786259541986e-06,
1955
+ "loss": 0.7837,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 0.7172236503856041,
1960
+ "grad_norm": 0.05332670360803604,
1961
+ "learning_rate": 9.585356528388456e-06,
1962
+ "loss": 0.7959,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 0.7197943444730077,
1967
+ "grad_norm": 0.05898366868495941,
1968
+ "learning_rate": 9.582918422060514e-06,
1969
+ "loss": 0.9204,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 0.7223650385604113,
1974
+ "grad_norm": 0.05881262570619583,
1975
+ "learning_rate": 9.580471897180128e-06,
1976
+ "loss": 0.8223,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 0.7249357326478149,
1981
+ "grad_norm": 0.05124802514910698,
1982
+ "learning_rate": 9.578016910069179e-06,
1983
+ "loss": 0.9062,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 0.7275064267352185,
1988
+ "grad_norm": 0.052421871572732925,
1989
+ "learning_rate": 9.575553416746873e-06,
1990
+ "loss": 0.8809,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 0.7300771208226221,
1995
+ "grad_norm": 0.05334588140249252,
1996
+ "learning_rate": 9.573081372927112e-06,
1997
+ "loss": 0.915,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 0.7326478149100257,
2002
+ "grad_norm": 0.05507427453994751,
2003
+ "learning_rate": 9.57060073401584e-06,
2004
+ "loss": 0.8315,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 0.7352185089974294,
2009
+ "grad_norm": 0.060719288885593414,
2010
+ "learning_rate": 9.56811145510836e-06,
2011
+ "loss": 0.812,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 0.7377892030848329,
2016
+ "grad_norm": 0.05697711929678917,
2017
+ "learning_rate": 9.565613490986625e-06,
2018
+ "loss": 0.9199,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 0.7403598971722365,
2023
+ "grad_norm": 0.05150134861469269,
2024
+ "learning_rate": 9.563106796116506e-06,
2025
+ "loss": 0.8496,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 0.7429305912596401,
2030
+ "grad_norm": 0.05363132804632187,
2031
+ "learning_rate": 9.56059132464501e-06,
2032
+ "loss": 0.7959,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 0.7455012853470437,
2037
+ "grad_norm": 0.05483046919107437,
2038
+ "learning_rate": 9.558067030397506e-06,
2039
+ "loss": 0.8789,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 0.7480719794344473,
2044
+ "grad_norm": 0.061016347259283066,
2045
+ "learning_rate": 9.555533866874878e-06,
2046
+ "loss": 1.0029,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 0.7506426735218509,
2051
+ "grad_norm": 0.058808423578739166,
2052
+ "learning_rate": 9.552991787250685e-06,
2053
+ "loss": 0.978,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 0.7532133676092545,
2058
+ "grad_norm": 0.05964820086956024,
2059
+ "learning_rate": 9.550440744368267e-06,
2060
+ "loss": 0.9038,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 0.7557840616966581,
2065
+ "grad_norm": 0.053899601101875305,
2066
+ "learning_rate": 9.547880690737834e-06,
2067
+ "loss": 0.8784,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 0.7583547557840618,
2072
+ "grad_norm": 0.05554712936282158,
2073
+ "learning_rate": 9.545311578533517e-06,
2074
+ "loss": 0.895,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 0.7609254498714653,
2079
+ "grad_norm": 0.05656404420733452,
2080
+ "learning_rate": 9.54273335959039e-06,
2081
+ "loss": 0.9814,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 0.7634961439588689,
2086
+ "grad_norm": 0.061445970088243484,
2087
+ "learning_rate": 9.54014598540146e-06,
2088
+ "loss": 0.8188,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 0.7660668380462725,
2093
+ "grad_norm": 0.05222717672586441,
2094
+ "learning_rate": 9.537549407114626e-06,
2095
+ "loss": 0.8276,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 0.7686375321336761,
2100
+ "grad_norm": 0.05029478669166565,
2101
+ "learning_rate": 9.534943575529599e-06,
2102
+ "loss": 0.8457,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 0.7712082262210797,
2107
+ "grad_norm": 0.055182084441185,
2108
+ "learning_rate": 9.532328441094805e-06,
2109
+ "loss": 1.0103,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 0.7737789203084833,
2114
+ "grad_norm": 0.051256079226732254,
2115
+ "learning_rate": 9.529703953904232e-06,
2116
+ "loss": 0.9023,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 0.7763496143958869,
2121
+ "grad_norm": 0.05435854196548462,
2122
+ "learning_rate": 9.527070063694269e-06,
2123
+ "loss": 0.895,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 0.7789203084832905,
2128
+ "grad_norm": 0.05296122655272484,
2129
+ "learning_rate": 9.524426719840478e-06,
2130
+ "loss": 0.9106,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 0.781491002570694,
2135
+ "grad_norm": 0.05457408353686333,
2136
+ "learning_rate": 9.521773871354376e-06,
2137
+ "loss": 0.9546,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 0.7840616966580977,
2142
+ "grad_norm": 0.0539555624127388,
2143
+ "learning_rate": 9.519111466880127e-06,
2144
+ "loss": 0.8506,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 0.7866323907455013,
2149
+ "grad_norm": 0.05563870817422867,
2150
+ "learning_rate": 9.51643945469126e-06,
2151
+ "loss": 0.9604,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 0.7892030848329049,
2156
+ "grad_norm": 0.05861511453986168,
2157
+ "learning_rate": 9.513757782687287e-06,
2158
+ "loss": 0.7812,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 0.7917737789203085,
2163
+ "grad_norm": 0.05230727419257164,
2164
+ "learning_rate": 9.511066398390344e-06,
2165
+ "loss": 0.8594,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 0.794344473007712,
2170
+ "grad_norm": 0.05595692992210388,
2171
+ "learning_rate": 9.508365248941746e-06,
2172
+ "loss": 0.9385,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 0.7969151670951157,
2177
+ "grad_norm": 0.05891738831996918,
2178
+ "learning_rate": 9.505654281098548e-06,
2179
+ "loss": 0.8284,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 0.7994858611825193,
2184
+ "grad_norm": 0.05573743209242821,
2185
+ "learning_rate": 9.502933441230022e-06,
2186
+ "loss": 0.9092,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 0.8020565552699229,
2191
+ "grad_norm": 0.05030527338385582,
2192
+ "learning_rate": 9.500202675314149e-06,
2193
+ "loss": 0.8823,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 0.8046272493573264,
2198
+ "grad_norm": 0.08709961920976639,
2199
+ "learning_rate": 9.497461928934011e-06,
2200
+ "loss": 0.8071,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 0.8071979434447301,
2205
+ "grad_norm": 0.10105069726705551,
2206
+ "learning_rate": 9.494711147274209e-06,
2207
+ "loss": 0.8877,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 0.8097686375321337,
2212
+ "grad_norm": 0.054532554000616074,
2213
+ "learning_rate": 9.49195027511718e-06,
2214
+ "loss": 0.8984,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 0.8123393316195373,
2219
+ "grad_norm": 0.05196183919906616,
2220
+ "learning_rate": 9.489179256839527e-06,
2221
+ "loss": 0.9268,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 0.8149100257069408,
2226
+ "grad_norm": 0.054404161870479584,
2227
+ "learning_rate": 9.486398036408264e-06,
2228
+ "loss": 0.9834,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 0.8174807197943444,
2233
+ "grad_norm": 0.051278892904520035,
2234
+ "learning_rate": 9.483606557377051e-06,
2235
+ "loss": 0.8032,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 0.8200514138817481,
2240
+ "grad_norm": 0.05271079018712044,
2241
+ "learning_rate": 9.480804762882367e-06,
2242
+ "loss": 0.8359,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 0.8226221079691517,
2247
+ "grad_norm": 0.05727388709783554,
2248
+ "learning_rate": 9.477992595639656e-06,
2249
+ "loss": 0.8398,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 0.8251928020565553,
2254
+ "grad_norm": 0.056882213801145554,
2255
+ "learning_rate": 9.47516999793942e-06,
2256
+ "loss": 0.8169,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 0.8277634961439588,
2261
+ "grad_norm": 0.05790659040212631,
2262
+ "learning_rate": 9.472336911643271e-06,
2263
+ "loss": 0.874,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 0.8303341902313625,
2268
+ "grad_norm": 0.052441854029893875,
2269
+ "learning_rate": 9.469493278179939e-06,
2270
+ "loss": 0.8394,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 0.8329048843187661,
2275
+ "grad_norm": 0.05015621706843376,
2276
+ "learning_rate": 9.466639038541236e-06,
2277
+ "loss": 0.8447,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 0.8354755784061697,
2282
+ "grad_norm": 0.06067253649234772,
2283
+ "learning_rate": 9.463774133277976e-06,
2284
+ "loss": 0.8623,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 0.8380462724935732,
2289
+ "grad_norm": 0.05550958961248398,
2290
+ "learning_rate": 9.46089850249584e-06,
2291
+ "loss": 0.9023,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 0.8406169665809768,
2296
+ "grad_norm": 0.05439194291830063,
2297
+ "learning_rate": 9.45801208585122e-06,
2298
+ "loss": 0.8516,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 0.8431876606683805,
2303
+ "grad_norm": 0.051754411309957504,
2304
+ "learning_rate": 9.455114822546972e-06,
2305
+ "loss": 0.8628,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 0.8457583547557841,
2310
+ "grad_norm": 0.05375438183546066,
2311
+ "learning_rate": 9.452206651328176e-06,
2312
+ "loss": 0.8701,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 0.8483290488431876,
2317
+ "grad_norm": 0.05142388865351677,
2318
+ "learning_rate": 9.449287510477788e-06,
2319
+ "loss": 0.9287,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 0.8508997429305912,
2324
+ "grad_norm": 0.054648611694574356,
2325
+ "learning_rate": 9.446357337812303e-06,
2326
+ "loss": 0.8638,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 0.8534704370179949,
2331
+ "grad_norm": 0.053628887981176376,
2332
+ "learning_rate": 9.443416070677324e-06,
2333
+ "loss": 0.8325,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 0.8560411311053985,
2338
+ "grad_norm": 0.05632541701197624,
2339
+ "learning_rate": 9.440463645943098e-06,
2340
+ "loss": 0.8271,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 0.8586118251928021,
2345
+ "grad_norm": 0.05134792625904083,
2346
+ "learning_rate": 9.4375e-06,
2347
+ "loss": 0.8179,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 0.8611825192802056,
2352
+ "grad_norm": 0.06385616213083267,
2353
+ "learning_rate": 9.434525068753969e-06,
2354
+ "loss": 0.7827,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 0.8637532133676092,
2359
+ "grad_norm": 0.054239194840192795,
2360
+ "learning_rate": 9.431538787621873e-06,
2361
+ "loss": 0.9009,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 0.8663239074550129,
2366
+ "grad_norm": 0.05689682811498642,
2367
+ "learning_rate": 9.428541091526865e-06,
2368
+ "loss": 0.8677,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 0.8688946015424165,
2373
+ "grad_norm": 0.05371352657675743,
2374
+ "learning_rate": 9.425531914893617e-06,
2375
+ "loss": 0.9834,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 0.87146529562982,
2380
+ "grad_norm": 0.054534927010536194,
2381
+ "learning_rate": 9.422511191643574e-06,
2382
+ "loss": 0.8618,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 0.8740359897172236,
2387
+ "grad_norm": 0.0579465813934803,
2388
+ "learning_rate": 9.41947885519009e-06,
2389
+ "loss": 0.8364,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 0.8766066838046273,
2394
+ "grad_norm": 0.04870595410466194,
2395
+ "learning_rate": 9.416434838433555e-06,
2396
+ "loss": 0.8594,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 0.8791773778920309,
2401
+ "grad_norm": 0.057385556399822235,
2402
+ "learning_rate": 9.413379073756432e-06,
2403
+ "loss": 0.8442,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 0.8817480719794345,
2408
+ "grad_norm": 0.07567571103572845,
2409
+ "learning_rate": 9.410311493018262e-06,
2410
+ "loss": 0.9521,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 0.884318766066838,
2415
+ "grad_norm": 0.0651511698961258,
2416
+ "learning_rate": 9.407232027550581e-06,
2417
+ "loss": 0.9077,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 0.8868894601542416,
2422
+ "grad_norm": 0.05570966750383377,
2423
+ "learning_rate": 9.404140608151824e-06,
2424
+ "loss": 0.9414,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 0.8894601542416453,
2429
+ "grad_norm": 0.05844790115952492,
2430
+ "learning_rate": 9.40103716508211e-06,
2431
+ "loss": 0.8198,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 0.8920308483290489,
2436
+ "grad_norm": 0.05628007650375366,
2437
+ "learning_rate": 9.397921628058021e-06,
2438
+ "loss": 0.9131,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 0.8946015424164524,
2443
+ "grad_norm": 0.05682618170976639,
2444
+ "learning_rate": 9.394793926247288e-06,
2445
+ "loss": 0.9604,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 0.897172236503856,
2450
+ "grad_norm": 0.05488917976617813,
2451
+ "learning_rate": 9.391653988263423e-06,
2452
+ "loss": 0.9189,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 0.8997429305912596,
2457
+ "grad_norm": 0.06217147782444954,
2458
+ "learning_rate": 9.38850174216028e-06,
2459
+ "loss": 0.782,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 0.9023136246786633,
2464
+ "grad_norm": 0.04932795837521553,
2465
+ "learning_rate": 9.385337115426577e-06,
2466
+ "loss": 0.7764,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 0.9048843187660668,
2471
+ "grad_norm": 0.056303396821022034,
2472
+ "learning_rate": 9.382160034980325e-06,
2473
+ "loss": 0.8872,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 0.9074550128534704,
2478
+ "grad_norm": 0.05455457419157028,
2479
+ "learning_rate": 9.3789704271632e-06,
2480
+ "loss": 0.7891,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 0.910025706940874,
2485
+ "grad_norm": 0.05396741256117821,
2486
+ "learning_rate": 9.375768217734856e-06,
2487
+ "loss": 0.9321,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 0.9125964010282777,
2492
+ "grad_norm": 0.0583309531211853,
2493
+ "learning_rate": 9.372553331867167e-06,
2494
+ "loss": 0.8823,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 0.9151670951156813,
2499
+ "grad_norm": 0.057046517729759216,
2500
+ "learning_rate": 9.369325694138386e-06,
2501
+ "loss": 0.9805,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 0.9177377892030848,
2506
+ "grad_norm": 0.05479217320680618,
2507
+ "learning_rate": 9.36608522852727e-06,
2508
+ "loss": 0.8901,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 0.9203084832904884,
2513
+ "grad_norm": 0.0561579167842865,
2514
+ "learning_rate": 9.36283185840708e-06,
2515
+ "loss": 0.8428,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 0.922879177377892,
2520
+ "grad_norm": 0.05958212539553642,
2521
+ "learning_rate": 9.35956550653957e-06,
2522
+ "loss": 1.0117,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 0.9254498714652957,
2527
+ "grad_norm": 0.05765485391020775,
2528
+ "learning_rate": 9.356286095068858e-06,
2529
+ "loss": 0.8096,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 0.9280205655526992,
2534
+ "grad_norm": 0.05282951146364212,
2535
+ "learning_rate": 9.352993545515247e-06,
2536
+ "loss": 0.813,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 0.9305912596401028,
2541
+ "grad_norm": 0.05436195805668831,
2542
+ "learning_rate": 9.349687778768957e-06,
2543
+ "loss": 0.835,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 0.9331619537275064,
2548
+ "grad_norm": 0.05730157718062401,
2549
+ "learning_rate": 9.346368715083801e-06,
2550
+ "loss": 0.7751,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 0.9357326478149101,
2555
+ "grad_norm": 0.059381358325481415,
2556
+ "learning_rate": 9.343036274070756e-06,
2557
+ "loss": 0.8716,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 0.9383033419023136,
2562
+ "grad_norm": 0.057895347476005554,
2563
+ "learning_rate": 9.339690374691496e-06,
2564
+ "loss": 0.8506,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 0.9408740359897172,
2569
+ "grad_norm": 0.052358921617269516,
2570
+ "learning_rate": 9.336330935251799e-06,
2571
+ "loss": 0.9102,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 0.9434447300771208,
2576
+ "grad_norm": 0.060520391911268234,
2577
+ "learning_rate": 9.332957873394909e-06,
2578
+ "loss": 0.9258,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 0.9460154241645244,
2583
+ "grad_norm": 0.055070772767066956,
2584
+ "learning_rate": 9.329571106094809e-06,
2585
+ "loss": 0.9722,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 0.9485861182519281,
2590
+ "grad_norm": 0.053908202797174454,
2591
+ "learning_rate": 9.3261705496494e-06,
2592
+ "loss": 0.8896,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 0.9511568123393316,
2597
+ "grad_norm": 0.057451847940683365,
2598
+ "learning_rate": 9.322756119673619e-06,
2599
+ "loss": 0.8467,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 0.9537275064267352,
2604
+ "grad_norm": 0.05375353991985321,
2605
+ "learning_rate": 9.319327731092437e-06,
2606
+ "loss": 0.8037,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 0.9562982005141388,
2611
+ "grad_norm": 0.05543424189090729,
2612
+ "learning_rate": 9.31588529813382e-06,
2613
+ "loss": 0.7646,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 0.9588688946015425,
2618
+ "grad_norm": 0.05233384296298027,
2619
+ "learning_rate": 9.31242873432155e-06,
2620
+ "loss": 0.8687,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 0.961439588688946,
2625
+ "grad_norm": 0.050029926002025604,
2626
+ "learning_rate": 9.308957952468008e-06,
2627
+ "loss": 0.8315,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 0.9640102827763496,
2632
+ "grad_norm": 0.07100562006235123,
2633
+ "learning_rate": 9.30547286466682e-06,
2634
+ "loss": 0.8276,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 0.9665809768637532,
2639
+ "grad_norm": 0.05098907649517059,
2640
+ "learning_rate": 9.301973382285454e-06,
2641
+ "loss": 0.9155,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 0.9691516709511568,
2646
+ "grad_norm": 0.04964357987046242,
2647
+ "learning_rate": 9.298459415957691e-06,
2648
+ "loss": 0.7583,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 0.9717223650385605,
2653
+ "grad_norm": 0.05745205283164978,
2654
+ "learning_rate": 9.294930875576038e-06,
2655
+ "loss": 0.9585,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 0.974293059125964,
2660
+ "grad_norm": 0.07048569619655609,
2661
+ "learning_rate": 9.291387670284e-06,
2662
+ "loss": 0.8726,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 0.9768637532133676,
2667
+ "grad_norm": 0.06114361807703972,
2668
+ "learning_rate": 9.287829708468302e-06,
2669
+ "loss": 0.8765,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 0.9794344473007712,
2674
+ "grad_norm": 0.052654024213552475,
2675
+ "learning_rate": 9.284256897750985e-06,
2676
+ "loss": 0.8281,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 0.9820051413881749,
2681
+ "grad_norm": 0.05415409430861473,
2682
+ "learning_rate": 9.280669144981413e-06,
2683
+ "loss": 0.8589,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 0.9845758354755784,
2688
+ "grad_norm": 0.05175313353538513,
2689
+ "learning_rate": 9.277066356228173e-06,
2690
+ "loss": 0.9443,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 0.987146529562982,
2695
+ "grad_norm": 0.05429228022694588,
2696
+ "learning_rate": 9.273448436770884e-06,
2697
+ "loss": 0.873,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 0.9897172236503856,
2702
+ "grad_norm": 0.05616312101483345,
2703
+ "learning_rate": 9.269815291091886e-06,
2704
+ "loss": 0.8818,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 0.9922879177377892,
2709
+ "grad_norm": 0.057027895003557205,
2710
+ "learning_rate": 9.266166822867854e-06,
2711
+ "loss": 0.8325,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 0.9948586118251928,
2716
+ "grad_norm": 0.0524408258497715,
2717
+ "learning_rate": 9.26250293496126e-06,
2718
+ "loss": 0.7954,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 0.9974293059125964,
2723
+ "grad_norm": 0.058645591139793396,
2724
+ "learning_rate": 9.258823529411765e-06,
2725
+ "loss": 0.8589,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.0,
2730
+ "grad_norm": 0.057895876467227936,
2731
+ "learning_rate": 9.255128507427495e-06,
2732
+ "loss": 0.8862,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.0025706940874035,
2737
+ "grad_norm": 0.04984109103679657,
2738
+ "learning_rate": 9.251417769376181e-06,
2739
+ "loss": 0.8076,
2740
+ "step": 390
2741
+ }
2742
+ ],
2743
+ "logging_steps": 1,
2744
+ "max_steps": 778,
2745
+ "num_input_tokens_seen": 0,
2746
+ "num_train_epochs": 2,
2747
+ "save_steps": 195,
2748
+ "stateful_callbacks": {
2749
+ "TrainerControl": {
2750
+ "args": {
2751
+ "should_epoch_stop": false,
2752
+ "should_evaluate": false,
2753
+ "should_log": false,
2754
+ "should_save": true,
2755
+ "should_training_stop": false
2756
+ },
2757
+ "attributes": {}
2758
+ }
2759
+ },
2760
+ "total_flos": 1.1343631119821046e+19,
2761
+ "train_batch_size": 2,
2762
+ "trial_name": null,
2763
+ "trial_params": null
2764
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff