SeungHeeKim commited on
Commit
42b0549
·
verified ·
1 Parent(s): f5f1fa6

Upload checkpoint-3930 to mmdocteam

Browse files
README.md ADDED
@@ -0,0 +1,206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen3-Embedding-4B
3
+ library_name: peft
4
+ tags:
5
+ - base_model:adapter:Qwen/Qwen3-Embedding-4B
6
+ - lora
7
+ - transformers
8
+ ---
9
+
10
+ # Model Card for Model ID
11
+
12
+ <!-- Provide a quick summary of what the model is/does. -->
13
+
14
+
15
+
16
+ ## Model Details
17
+
18
+ ### Model Description
19
+
20
+ <!-- Provide a longer summary of what this model is. -->
21
+
22
+
23
+
24
+ - **Developed by:** [More Information Needed]
25
+ - **Funded by [optional]:** [More Information Needed]
26
+ - **Shared by [optional]:** [More Information Needed]
27
+ - **Model type:** [More Information Needed]
28
+ - **Language(s) (NLP):** [More Information Needed]
29
+ - **License:** [More Information Needed]
30
+ - **Finetuned from model [optional]:** [More Information Needed]
31
+
32
+ ### Model Sources [optional]
33
+
34
+ <!-- Provide the basic links for the model. -->
35
+
36
+ - **Repository:** [More Information Needed]
37
+ - **Paper [optional]:** [More Information Needed]
38
+ - **Demo [optional]:** [More Information Needed]
39
+
40
+ ## Uses
41
+
42
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
43
+
44
+ ### Direct Use
45
+
46
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
47
+
48
+ [More Information Needed]
49
+
50
+ ### Downstream Use [optional]
51
+
52
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
53
+
54
+ [More Information Needed]
55
+
56
+ ### Out-of-Scope Use
57
+
58
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
59
+
60
+ [More Information Needed]
61
+
62
+ ## Bias, Risks, and Limitations
63
+
64
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
65
+
66
+ [More Information Needed]
67
+
68
+ ### Recommendations
69
+
70
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
71
+
72
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
73
+
74
+ ## How to Get Started with the Model
75
+
76
+ Use the code below to get started with the model.
77
+
78
+ [More Information Needed]
79
+
80
+ ## Training Details
81
+
82
+ ### Training Data
83
+
84
+ <!-- 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. -->
85
+
86
+ [More Information Needed]
87
+
88
+ ### Training Procedure
89
+
90
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
91
+
92
+ #### Preprocessing [optional]
93
+
94
+ [More Information Needed]
95
+
96
+
97
+ #### Training Hyperparameters
98
+
99
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
100
+
101
+ #### Speeds, Sizes, Times [optional]
102
+
103
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
104
+
105
+ [More Information Needed]
106
+
107
+ ## Evaluation
108
+
109
+ <!-- This section describes the evaluation protocols and provides the results. -->
110
+
111
+ ### Testing Data, Factors & Metrics
112
+
113
+ #### Testing Data
114
+
115
+ <!-- This should link to a Dataset Card if possible. -->
116
+
117
+ [More Information Needed]
118
+
119
+ #### Factors
120
+
121
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
122
+
123
+ [More Information Needed]
124
+
125
+ #### Metrics
126
+
127
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
128
+
129
+ [More Information Needed]
130
+
131
+ ### Results
132
+
133
+ [More Information Needed]
134
+
135
+ #### Summary
136
+
137
+
138
+
139
+ ## Model Examination [optional]
140
+
141
+ <!-- Relevant interpretability work for the model goes here -->
142
+
143
+ [More Information Needed]
144
+
145
+ ## Environmental Impact
146
+
147
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
148
+
149
+ 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).
150
+
151
+ - **Hardware Type:** [More Information Needed]
152
+ - **Hours used:** [More Information Needed]
153
+ - **Cloud Provider:** [More Information Needed]
154
+ - **Compute Region:** [More Information Needed]
155
+ - **Carbon Emitted:** [More Information Needed]
156
+
157
+ ## Technical Specifications [optional]
158
+
159
+ ### Model Architecture and Objective
160
+
161
+ [More Information Needed]
162
+
163
+ ### Compute Infrastructure
164
+
165
+ [More Information Needed]
166
+
167
+ #### Hardware
168
+
169
+ [More Information Needed]
170
+
171
+ #### Software
172
+
173
+ [More Information Needed]
174
+
175
+ ## Citation [optional]
176
+
177
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
178
+
179
+ **BibTeX:**
180
+
181
+ [More Information Needed]
182
+
183
+ **APA:**
184
+
185
+ [More Information Needed]
186
+
187
+ ## Glossary [optional]
188
+
189
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
190
+
191
+ [More Information Needed]
192
+
193
+ ## More Information [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Authors [optional]
198
+
199
+ [More Information Needed]
200
+
201
+ ## Model Card Contact
202
+
203
+ [More Information Needed]
204
+ ### Framework versions
205
+
206
+ - PEFT 0.17.1
adapter_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Qwen/Qwen3-Embedding-4B",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": "gaussian",
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 32,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0.05,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "qalora_group_size": 16,
24
+ "r": 8,
25
+ "rank_pattern": {},
26
+ "revision": null,
27
+ "target_modules": [
28
+ "custom_text_proj",
29
+ "v_proj",
30
+ "q_proj",
31
+ "k_proj",
32
+ "o_proj"
33
+ ],
34
+ "target_parameters": null,
35
+ "task_type": "FEATURE_EXTRACTION",
36
+ "trainable_token_indices": null,
37
+ "use_dora": false,
38
+ "use_qalora": false,
39
+ "use_rslora": false
40
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0391acb614994855d53ec94747db4d52661b05e1b8966733119612b125029257
3
+ size 23715752
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f67647422d7b9e1eed6b739d995bef27d5df9936d88e2431d215d0926b61a6a5
3
+ size 47605771
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7157f9a8289c934ee26cf20d44c97b41f367a18e76f31c6b02bc100041641972
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9401f1ba85746e94b06746d4be4f66667a6cb981229a822a85bd4d383a37ba9
3
+ size 1465
trainer_state.json ADDED
@@ -0,0 +1,2540 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.8905852417302799,
6
+ "eval_steps": 500,
7
+ "global_step": 3500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.002544529262086514,
14
+ "grad_norm": 8.495231628417969,
15
+ "learning_rate": 9.160305343511451e-07,
16
+ "loss": 0.1152,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.005089058524173028,
21
+ "grad_norm": 9.448420524597168,
22
+ "learning_rate": 1.933842239185751e-06,
23
+ "loss": 0.1589,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.007633587786259542,
28
+ "grad_norm": 12.052129745483398,
29
+ "learning_rate": 2.951653944020356e-06,
30
+ "loss": 0.1132,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.010178117048346057,
35
+ "grad_norm": 13.677345275878906,
36
+ "learning_rate": 3.969465648854962e-06,
37
+ "loss": 0.1104,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.01272264631043257,
42
+ "grad_norm": 9.209020614624023,
43
+ "learning_rate": 4.987277353689568e-06,
44
+ "loss": 0.0461,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.015267175572519083,
49
+ "grad_norm": 14.143465995788574,
50
+ "learning_rate": 6.005089058524174e-06,
51
+ "loss": 0.0482,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.017811704834605598,
56
+ "grad_norm": 1.9570101499557495,
57
+ "learning_rate": 7.022900763358779e-06,
58
+ "loss": 0.0331,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.020356234096692113,
63
+ "grad_norm": 6.45846700668335,
64
+ "learning_rate": 8.040712468193384e-06,
65
+ "loss": 0.0358,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.022900763358778626,
70
+ "grad_norm": 6.0744309425354,
71
+ "learning_rate": 9.058524173027991e-06,
72
+ "loss": 0.026,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.02544529262086514,
77
+ "grad_norm": 8.33021068572998,
78
+ "learning_rate": 1.0076335877862595e-05,
79
+ "loss": 0.0373,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.027989821882951654,
84
+ "grad_norm": 6.4917778968811035,
85
+ "learning_rate": 1.1094147582697202e-05,
86
+ "loss": 0.0186,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.030534351145038167,
91
+ "grad_norm": 4.850451469421387,
92
+ "learning_rate": 1.2111959287531807e-05,
93
+ "loss": 0.0191,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.03307888040712468,
98
+ "grad_norm": 3.528228521347046,
99
+ "learning_rate": 1.3129770992366414e-05,
100
+ "loss": 0.0244,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.035623409669211195,
105
+ "grad_norm": 0.09570707380771637,
106
+ "learning_rate": 1.4147582697201019e-05,
107
+ "loss": 0.0154,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.03816793893129771,
112
+ "grad_norm": 8.72260856628418,
113
+ "learning_rate": 1.5165394402035624e-05,
114
+ "loss": 0.0251,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.04071246819338423,
119
+ "grad_norm": 5.22550106048584,
120
+ "learning_rate": 1.618320610687023e-05,
121
+ "loss": 0.0181,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.043256997455470736,
126
+ "grad_norm": 7.037172317504883,
127
+ "learning_rate": 1.7201017811704836e-05,
128
+ "loss": 0.007,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.04580152671755725,
133
+ "grad_norm": 0.9109981060028076,
134
+ "learning_rate": 1.8218829516539443e-05,
135
+ "loss": 0.0177,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.04834605597964377,
140
+ "grad_norm": 6.822402477264404,
141
+ "learning_rate": 1.923664122137405e-05,
142
+ "loss": 0.0079,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.05089058524173028,
147
+ "grad_norm": 2.628634452819824,
148
+ "learning_rate": 2.0254452926208653e-05,
149
+ "loss": 0.0142,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.05343511450381679,
154
+ "grad_norm": 0.7327078580856323,
155
+ "learning_rate": 2.127226463104326e-05,
156
+ "loss": 0.0207,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.05597964376590331,
161
+ "grad_norm": 7.423736095428467,
162
+ "learning_rate": 2.2290076335877867e-05,
163
+ "loss": 0.0127,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.058524173027989825,
168
+ "grad_norm": 0.46404168009757996,
169
+ "learning_rate": 2.330788804071247e-05,
170
+ "loss": 0.0158,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.061068702290076333,
175
+ "grad_norm": 0.023200908675789833,
176
+ "learning_rate": 2.4325699745547078e-05,
177
+ "loss": 0.0178,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.06361323155216285,
182
+ "grad_norm": 3.8594086170196533,
183
+ "learning_rate": 2.5343511450381678e-05,
184
+ "loss": 0.0069,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.06615776081424936,
189
+ "grad_norm": 0.4054301679134369,
190
+ "learning_rate": 2.6361323155216285e-05,
191
+ "loss": 0.0223,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.06870229007633588,
196
+ "grad_norm": 2.2085139751434326,
197
+ "learning_rate": 2.737913486005089e-05,
198
+ "loss": 0.0362,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.07124681933842239,
203
+ "grad_norm": 0.5911164283752441,
204
+ "learning_rate": 2.8396946564885498e-05,
205
+ "loss": 0.0187,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.0737913486005089,
210
+ "grad_norm": 4.2948102951049805,
211
+ "learning_rate": 2.9414758269720102e-05,
212
+ "loss": 0.0107,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.07633587786259542,
217
+ "grad_norm": 0.2595175504684448,
218
+ "learning_rate": 3.043256997455471e-05,
219
+ "loss": 0.0085,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.07888040712468193,
224
+ "grad_norm": 5.613988876342773,
225
+ "learning_rate": 3.145038167938931e-05,
226
+ "loss": 0.0281,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.08142493638676845,
231
+ "grad_norm": 2.6039955615997314,
232
+ "learning_rate": 3.246819338422392e-05,
233
+ "loss": 0.0213,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.08396946564885496,
238
+ "grad_norm": 0.2858663499355316,
239
+ "learning_rate": 3.3486005089058526e-05,
240
+ "loss": 0.0141,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.08651399491094147,
245
+ "grad_norm": 0.19223034381866455,
246
+ "learning_rate": 3.450381679389313e-05,
247
+ "loss": 0.0224,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.089058524173028,
252
+ "grad_norm": 5.314587116241455,
253
+ "learning_rate": 3.552162849872774e-05,
254
+ "loss": 0.0129,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.0916030534351145,
259
+ "grad_norm": 4.426294326782227,
260
+ "learning_rate": 3.653944020356235e-05,
261
+ "loss": 0.0224,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.09414758269720101,
266
+ "grad_norm": 5.395482063293457,
267
+ "learning_rate": 3.755725190839695e-05,
268
+ "loss": 0.01,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.09669211195928754,
273
+ "grad_norm": 0.7296554446220398,
274
+ "learning_rate": 3.8575063613231554e-05,
275
+ "loss": 0.0192,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.09923664122137404,
280
+ "grad_norm": 0.2524012625217438,
281
+ "learning_rate": 3.959287531806616e-05,
282
+ "loss": 0.0178,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.10178117048346055,
287
+ "grad_norm": 5.55770206451416,
288
+ "learning_rate": 3.993214588634436e-05,
289
+ "loss": 0.0229,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.10432569974554708,
294
+ "grad_norm": 4.481260299682617,
295
+ "learning_rate": 3.9819055696918295e-05,
296
+ "loss": 0.0138,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.10687022900763359,
301
+ "grad_norm": 2.183645725250244,
302
+ "learning_rate": 3.970596550749223e-05,
303
+ "loss": 0.0201,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.10941475826972011,
308
+ "grad_norm": 3.418135404586792,
309
+ "learning_rate": 3.959287531806616e-05,
310
+ "loss": 0.0225,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.11195928753180662,
315
+ "grad_norm": 0.1539747565984726,
316
+ "learning_rate": 3.947978512864009e-05,
317
+ "loss": 0.0043,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.11450381679389313,
322
+ "grad_norm": 3.93865966796875,
323
+ "learning_rate": 3.9366694939214026e-05,
324
+ "loss": 0.0088,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.11704834605597965,
329
+ "grad_norm": 2.0105717182159424,
330
+ "learning_rate": 3.925360474978796e-05,
331
+ "loss": 0.0101,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.11959287531806616,
336
+ "grad_norm": 1.3878592252731323,
337
+ "learning_rate": 3.914051456036189e-05,
338
+ "loss": 0.0086,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.12213740458015267,
343
+ "grad_norm": 0.09708156436681747,
344
+ "learning_rate": 3.902742437093582e-05,
345
+ "loss": 0.0119,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.12468193384223919,
350
+ "grad_norm": 0.4155742824077606,
351
+ "learning_rate": 3.8914334181509756e-05,
352
+ "loss": 0.0139,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.1272264631043257,
357
+ "grad_norm": 0.6312187910079956,
358
+ "learning_rate": 3.880124399208369e-05,
359
+ "loss": 0.0185,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.1272264631043257,
364
+ "eval_loss": 0.0426296703517437,
365
+ "eval_runtime": 100.8006,
366
+ "eval_samples_per_second": 79.186,
367
+ "eval_steps_per_second": 0.625,
368
+ "step": 500
369
+ },
370
+ {
371
+ "epoch": 0.1297709923664122,
372
+ "grad_norm": 8.71293830871582,
373
+ "learning_rate": 3.868815380265762e-05,
374
+ "loss": 0.0236,
375
+ "step": 510
376
+ },
377
+ {
378
+ "epoch": 0.13231552162849872,
379
+ "grad_norm": 0.6096652150154114,
380
+ "learning_rate": 3.8575063613231554e-05,
381
+ "loss": 0.01,
382
+ "step": 520
383
+ },
384
+ {
385
+ "epoch": 0.13486005089058525,
386
+ "grad_norm": 0.9275781512260437,
387
+ "learning_rate": 3.8461973423805486e-05,
388
+ "loss": 0.0091,
389
+ "step": 530
390
+ },
391
+ {
392
+ "epoch": 0.13740458015267176,
393
+ "grad_norm": 0.3549460768699646,
394
+ "learning_rate": 3.834888323437942e-05,
395
+ "loss": 0.0092,
396
+ "step": 540
397
+ },
398
+ {
399
+ "epoch": 0.13994910941475827,
400
+ "grad_norm": 3.617748498916626,
401
+ "learning_rate": 3.823579304495336e-05,
402
+ "loss": 0.0117,
403
+ "step": 550
404
+ },
405
+ {
406
+ "epoch": 0.14249363867684478,
407
+ "grad_norm": 3.950561761856079,
408
+ "learning_rate": 3.8122702855527284e-05,
409
+ "loss": 0.0134,
410
+ "step": 560
411
+ },
412
+ {
413
+ "epoch": 0.1450381679389313,
414
+ "grad_norm": 1.9877593517303467,
415
+ "learning_rate": 3.8009612666101216e-05,
416
+ "loss": 0.0143,
417
+ "step": 570
418
+ },
419
+ {
420
+ "epoch": 0.1475826972010178,
421
+ "grad_norm": 1.1295750141143799,
422
+ "learning_rate": 3.789652247667515e-05,
423
+ "loss": 0.0035,
424
+ "step": 580
425
+ },
426
+ {
427
+ "epoch": 0.15012722646310434,
428
+ "grad_norm": 0.07594585418701172,
429
+ "learning_rate": 3.778343228724908e-05,
430
+ "loss": 0.011,
431
+ "step": 590
432
+ },
433
+ {
434
+ "epoch": 0.15267175572519084,
435
+ "grad_norm": 0.28959858417510986,
436
+ "learning_rate": 3.767034209782302e-05,
437
+ "loss": 0.0034,
438
+ "step": 600
439
+ },
440
+ {
441
+ "epoch": 0.15521628498727735,
442
+ "grad_norm": 0.31216809153556824,
443
+ "learning_rate": 3.755725190839695e-05,
444
+ "loss": 0.0145,
445
+ "step": 610
446
+ },
447
+ {
448
+ "epoch": 0.15776081424936386,
449
+ "grad_norm": 0.048563506454229355,
450
+ "learning_rate": 3.744416171897088e-05,
451
+ "loss": 0.0136,
452
+ "step": 620
453
+ },
454
+ {
455
+ "epoch": 0.16030534351145037,
456
+ "grad_norm": 0.18490168452262878,
457
+ "learning_rate": 3.733107152954481e-05,
458
+ "loss": 0.0157,
459
+ "step": 630
460
+ },
461
+ {
462
+ "epoch": 0.1628498727735369,
463
+ "grad_norm": 4.179055690765381,
464
+ "learning_rate": 3.721798134011875e-05,
465
+ "loss": 0.0113,
466
+ "step": 640
467
+ },
468
+ {
469
+ "epoch": 0.16539440203562342,
470
+ "grad_norm": 0.7711288332939148,
471
+ "learning_rate": 3.7104891150692684e-05,
472
+ "loss": 0.004,
473
+ "step": 650
474
+ },
475
+ {
476
+ "epoch": 0.16793893129770993,
477
+ "grad_norm": 2.4299933910369873,
478
+ "learning_rate": 3.699180096126661e-05,
479
+ "loss": 0.0093,
480
+ "step": 660
481
+ },
482
+ {
483
+ "epoch": 0.17048346055979643,
484
+ "grad_norm": 1.621607780456543,
485
+ "learning_rate": 3.687871077184054e-05,
486
+ "loss": 0.0071,
487
+ "step": 670
488
+ },
489
+ {
490
+ "epoch": 0.17302798982188294,
491
+ "grad_norm": 0.06137412041425705,
492
+ "learning_rate": 3.676562058241448e-05,
493
+ "loss": 0.0216,
494
+ "step": 680
495
+ },
496
+ {
497
+ "epoch": 0.17557251908396945,
498
+ "grad_norm": 1.6078226566314697,
499
+ "learning_rate": 3.6652530392988414e-05,
500
+ "loss": 0.0242,
501
+ "step": 690
502
+ },
503
+ {
504
+ "epoch": 0.178117048346056,
505
+ "grad_norm": 2.2799558639526367,
506
+ "learning_rate": 3.653944020356235e-05,
507
+ "loss": 0.0113,
508
+ "step": 700
509
+ },
510
+ {
511
+ "epoch": 0.1806615776081425,
512
+ "grad_norm": 2.2463083267211914,
513
+ "learning_rate": 3.642635001413627e-05,
514
+ "loss": 0.0054,
515
+ "step": 710
516
+ },
517
+ {
518
+ "epoch": 0.183206106870229,
519
+ "grad_norm": 0.4119090437889099,
520
+ "learning_rate": 3.6313259824710205e-05,
521
+ "loss": 0.0107,
522
+ "step": 720
523
+ },
524
+ {
525
+ "epoch": 0.18575063613231552,
526
+ "grad_norm": 3.735358476638794,
527
+ "learning_rate": 3.6200169635284144e-05,
528
+ "loss": 0.0096,
529
+ "step": 730
530
+ },
531
+ {
532
+ "epoch": 0.18829516539440203,
533
+ "grad_norm": 2.4470431804656982,
534
+ "learning_rate": 3.608707944585808e-05,
535
+ "loss": 0.0089,
536
+ "step": 740
537
+ },
538
+ {
539
+ "epoch": 0.19083969465648856,
540
+ "grad_norm": 0.0336996465921402,
541
+ "learning_rate": 3.597398925643201e-05,
542
+ "loss": 0.0111,
543
+ "step": 750
544
+ },
545
+ {
546
+ "epoch": 0.19338422391857507,
547
+ "grad_norm": 0.6565234661102295,
548
+ "learning_rate": 3.586089906700594e-05,
549
+ "loss": 0.0083,
550
+ "step": 760
551
+ },
552
+ {
553
+ "epoch": 0.19592875318066158,
554
+ "grad_norm": 9.512853622436523,
555
+ "learning_rate": 3.5747808877579875e-05,
556
+ "loss": 0.0051,
557
+ "step": 770
558
+ },
559
+ {
560
+ "epoch": 0.1984732824427481,
561
+ "grad_norm": 0.5155350565910339,
562
+ "learning_rate": 3.563471868815381e-05,
563
+ "loss": 0.008,
564
+ "step": 780
565
+ },
566
+ {
567
+ "epoch": 0.2010178117048346,
568
+ "grad_norm": 4.279949188232422,
569
+ "learning_rate": 3.552162849872774e-05,
570
+ "loss": 0.0081,
571
+ "step": 790
572
+ },
573
+ {
574
+ "epoch": 0.2035623409669211,
575
+ "grad_norm": 2.585608959197998,
576
+ "learning_rate": 3.540853830930167e-05,
577
+ "loss": 0.018,
578
+ "step": 800
579
+ },
580
+ {
581
+ "epoch": 0.20610687022900764,
582
+ "grad_norm": 1.0371241569519043,
583
+ "learning_rate": 3.5295448119875605e-05,
584
+ "loss": 0.0134,
585
+ "step": 810
586
+ },
587
+ {
588
+ "epoch": 0.20865139949109415,
589
+ "grad_norm": 0.7093348503112793,
590
+ "learning_rate": 3.518235793044954e-05,
591
+ "loss": 0.0035,
592
+ "step": 820
593
+ },
594
+ {
595
+ "epoch": 0.21119592875318066,
596
+ "grad_norm": 0.48215287923812866,
597
+ "learning_rate": 3.506926774102347e-05,
598
+ "loss": 0.0099,
599
+ "step": 830
600
+ },
601
+ {
602
+ "epoch": 0.21374045801526717,
603
+ "grad_norm": 3.264940023422241,
604
+ "learning_rate": 3.49561775515974e-05,
605
+ "loss": 0.0152,
606
+ "step": 840
607
+ },
608
+ {
609
+ "epoch": 0.21628498727735368,
610
+ "grad_norm": 0.03547951951622963,
611
+ "learning_rate": 3.4843087362171335e-05,
612
+ "loss": 0.0078,
613
+ "step": 850
614
+ },
615
+ {
616
+ "epoch": 0.21882951653944022,
617
+ "grad_norm": 2.6558687686920166,
618
+ "learning_rate": 3.472999717274527e-05,
619
+ "loss": 0.0134,
620
+ "step": 860
621
+ },
622
+ {
623
+ "epoch": 0.22137404580152673,
624
+ "grad_norm": 0.41831302642822266,
625
+ "learning_rate": 3.46169069833192e-05,
626
+ "loss": 0.0045,
627
+ "step": 870
628
+ },
629
+ {
630
+ "epoch": 0.22391857506361323,
631
+ "grad_norm": 0.04895065724849701,
632
+ "learning_rate": 3.450381679389313e-05,
633
+ "loss": 0.0093,
634
+ "step": 880
635
+ },
636
+ {
637
+ "epoch": 0.22646310432569974,
638
+ "grad_norm": 4.139157772064209,
639
+ "learning_rate": 3.4390726604467065e-05,
640
+ "loss": 0.0122,
641
+ "step": 890
642
+ },
643
+ {
644
+ "epoch": 0.22900763358778625,
645
+ "grad_norm": 0.1418493539094925,
646
+ "learning_rate": 3.4277636415041e-05,
647
+ "loss": 0.0147,
648
+ "step": 900
649
+ },
650
+ {
651
+ "epoch": 0.23155216284987276,
652
+ "grad_norm": 0.19460724294185638,
653
+ "learning_rate": 3.416454622561493e-05,
654
+ "loss": 0.0128,
655
+ "step": 910
656
+ },
657
+ {
658
+ "epoch": 0.2340966921119593,
659
+ "grad_norm": 0.18406054377555847,
660
+ "learning_rate": 3.405145603618886e-05,
661
+ "loss": 0.0115,
662
+ "step": 920
663
+ },
664
+ {
665
+ "epoch": 0.2366412213740458,
666
+ "grad_norm": 0.16981343924999237,
667
+ "learning_rate": 3.3938365846762796e-05,
668
+ "loss": 0.0082,
669
+ "step": 930
670
+ },
671
+ {
672
+ "epoch": 0.23918575063613232,
673
+ "grad_norm": 0.009775158949196339,
674
+ "learning_rate": 3.382527565733673e-05,
675
+ "loss": 0.0027,
676
+ "step": 940
677
+ },
678
+ {
679
+ "epoch": 0.24173027989821882,
680
+ "grad_norm": 0.1106642484664917,
681
+ "learning_rate": 3.371218546791066e-05,
682
+ "loss": 0.0139,
683
+ "step": 950
684
+ },
685
+ {
686
+ "epoch": 0.24427480916030533,
687
+ "grad_norm": 0.10489284992218018,
688
+ "learning_rate": 3.3599095278484593e-05,
689
+ "loss": 0.002,
690
+ "step": 960
691
+ },
692
+ {
693
+ "epoch": 0.24681933842239187,
694
+ "grad_norm": 0.26976943016052246,
695
+ "learning_rate": 3.3486005089058526e-05,
696
+ "loss": 0.0075,
697
+ "step": 970
698
+ },
699
+ {
700
+ "epoch": 0.24936386768447838,
701
+ "grad_norm": 0.06751968711614609,
702
+ "learning_rate": 3.337291489963246e-05,
703
+ "loss": 0.015,
704
+ "step": 980
705
+ },
706
+ {
707
+ "epoch": 0.25190839694656486,
708
+ "grad_norm": 0.07933365553617477,
709
+ "learning_rate": 3.325982471020639e-05,
710
+ "loss": 0.0049,
711
+ "step": 990
712
+ },
713
+ {
714
+ "epoch": 0.2544529262086514,
715
+ "grad_norm": 1.002841830253601,
716
+ "learning_rate": 3.3146734520780324e-05,
717
+ "loss": 0.0082,
718
+ "step": 1000
719
+ },
720
+ {
721
+ "epoch": 0.2544529262086514,
722
+ "eval_loss": 0.04029834270477295,
723
+ "eval_runtime": 100.7792,
724
+ "eval_samples_per_second": 79.203,
725
+ "eval_steps_per_second": 0.625,
726
+ "step": 1000
727
+ },
728
+ {
729
+ "epoch": 0.25699745547073793,
730
+ "grad_norm": 0.11880303919315338,
731
+ "learning_rate": 3.3033644331354256e-05,
732
+ "loss": 0.0071,
733
+ "step": 1010
734
+ },
735
+ {
736
+ "epoch": 0.2595419847328244,
737
+ "grad_norm": 2.1375386714935303,
738
+ "learning_rate": 3.292055414192819e-05,
739
+ "loss": 0.0043,
740
+ "step": 1020
741
+ },
742
+ {
743
+ "epoch": 0.26208651399491095,
744
+ "grad_norm": 1.9007747173309326,
745
+ "learning_rate": 3.280746395250212e-05,
746
+ "loss": 0.0056,
747
+ "step": 1030
748
+ },
749
+ {
750
+ "epoch": 0.26463104325699743,
751
+ "grad_norm": 1.3233751058578491,
752
+ "learning_rate": 3.2694373763076054e-05,
753
+ "loss": 0.0122,
754
+ "step": 1040
755
+ },
756
+ {
757
+ "epoch": 0.26717557251908397,
758
+ "grad_norm": 1.3246195316314697,
759
+ "learning_rate": 3.2581283573649987e-05,
760
+ "loss": 0.0082,
761
+ "step": 1050
762
+ },
763
+ {
764
+ "epoch": 0.2697201017811705,
765
+ "grad_norm": 3.8455071449279785,
766
+ "learning_rate": 3.246819338422392e-05,
767
+ "loss": 0.0048,
768
+ "step": 1060
769
+ },
770
+ {
771
+ "epoch": 0.272264631043257,
772
+ "grad_norm": 1.4055747985839844,
773
+ "learning_rate": 3.235510319479785e-05,
774
+ "loss": 0.008,
775
+ "step": 1070
776
+ },
777
+ {
778
+ "epoch": 0.2748091603053435,
779
+ "grad_norm": 1.1376632452011108,
780
+ "learning_rate": 3.224201300537179e-05,
781
+ "loss": 0.0088,
782
+ "step": 1080
783
+ },
784
+ {
785
+ "epoch": 0.27735368956743,
786
+ "grad_norm": 1.465824007987976,
787
+ "learning_rate": 3.212892281594572e-05,
788
+ "loss": 0.0082,
789
+ "step": 1090
790
+ },
791
+ {
792
+ "epoch": 0.27989821882951654,
793
+ "grad_norm": 0.5838302373886108,
794
+ "learning_rate": 3.201583262651965e-05,
795
+ "loss": 0.0128,
796
+ "step": 1100
797
+ },
798
+ {
799
+ "epoch": 0.2824427480916031,
800
+ "grad_norm": 0.008899681270122528,
801
+ "learning_rate": 3.190274243709358e-05,
802
+ "loss": 0.0061,
803
+ "step": 1110
804
+ },
805
+ {
806
+ "epoch": 0.28498727735368956,
807
+ "grad_norm": 2.9608395099639893,
808
+ "learning_rate": 3.178965224766752e-05,
809
+ "loss": 0.0217,
810
+ "step": 1120
811
+ },
812
+ {
813
+ "epoch": 0.2875318066157761,
814
+ "grad_norm": 0.12024850398302078,
815
+ "learning_rate": 3.1676562058241454e-05,
816
+ "loss": 0.0072,
817
+ "step": 1130
818
+ },
819
+ {
820
+ "epoch": 0.2900763358778626,
821
+ "grad_norm": 0.09325462579727173,
822
+ "learning_rate": 3.156347186881538e-05,
823
+ "loss": 0.0079,
824
+ "step": 1140
825
+ },
826
+ {
827
+ "epoch": 0.2926208651399491,
828
+ "grad_norm": 0.047860078513622284,
829
+ "learning_rate": 3.145038167938931e-05,
830
+ "loss": 0.0111,
831
+ "step": 1150
832
+ },
833
+ {
834
+ "epoch": 0.2951653944020356,
835
+ "grad_norm": 0.14491482079029083,
836
+ "learning_rate": 3.1337291489963245e-05,
837
+ "loss": 0.0079,
838
+ "step": 1160
839
+ },
840
+ {
841
+ "epoch": 0.29770992366412213,
842
+ "grad_norm": 5.476013660430908,
843
+ "learning_rate": 3.1224201300537184e-05,
844
+ "loss": 0.0161,
845
+ "step": 1170
846
+ },
847
+ {
848
+ "epoch": 0.30025445292620867,
849
+ "grad_norm": 0.5506300926208496,
850
+ "learning_rate": 3.111111111111112e-05,
851
+ "loss": 0.0032,
852
+ "step": 1180
853
+ },
854
+ {
855
+ "epoch": 0.30279898218829515,
856
+ "grad_norm": 0.5853263735771179,
857
+ "learning_rate": 3.099802092168504e-05,
858
+ "loss": 0.0107,
859
+ "step": 1190
860
+ },
861
+ {
862
+ "epoch": 0.3053435114503817,
863
+ "grad_norm": 0.47357645630836487,
864
+ "learning_rate": 3.0884930732258975e-05,
865
+ "loss": 0.0046,
866
+ "step": 1200
867
+ },
868
+ {
869
+ "epoch": 0.30788804071246817,
870
+ "grad_norm": 0.004211845807731152,
871
+ "learning_rate": 3.0771840542832914e-05,
872
+ "loss": 0.0099,
873
+ "step": 1210
874
+ },
875
+ {
876
+ "epoch": 0.3104325699745547,
877
+ "grad_norm": 0.029812565073370934,
878
+ "learning_rate": 3.065875035340685e-05,
879
+ "loss": 0.0094,
880
+ "step": 1220
881
+ },
882
+ {
883
+ "epoch": 0.31297709923664124,
884
+ "grad_norm": 2.437438488006592,
885
+ "learning_rate": 3.054566016398078e-05,
886
+ "loss": 0.014,
887
+ "step": 1230
888
+ },
889
+ {
890
+ "epoch": 0.3155216284987277,
891
+ "grad_norm": 0.3200075626373291,
892
+ "learning_rate": 3.043256997455471e-05,
893
+ "loss": 0.0082,
894
+ "step": 1240
895
+ },
896
+ {
897
+ "epoch": 0.31806615776081426,
898
+ "grad_norm": 0.023235173895955086,
899
+ "learning_rate": 3.031947978512864e-05,
900
+ "loss": 0.013,
901
+ "step": 1250
902
+ },
903
+ {
904
+ "epoch": 0.32061068702290074,
905
+ "grad_norm": 0.042968690395355225,
906
+ "learning_rate": 3.0206389595702577e-05,
907
+ "loss": 0.0222,
908
+ "step": 1260
909
+ },
910
+ {
911
+ "epoch": 0.3231552162849873,
912
+ "grad_norm": 0.07905975729227066,
913
+ "learning_rate": 3.009329940627651e-05,
914
+ "loss": 0.004,
915
+ "step": 1270
916
+ },
917
+ {
918
+ "epoch": 0.3256997455470738,
919
+ "grad_norm": 0.47882354259490967,
920
+ "learning_rate": 2.998020921685044e-05,
921
+ "loss": 0.0042,
922
+ "step": 1280
923
+ },
924
+ {
925
+ "epoch": 0.3282442748091603,
926
+ "grad_norm": 1.036494255065918,
927
+ "learning_rate": 2.986711902742437e-05,
928
+ "loss": 0.0032,
929
+ "step": 1290
930
+ },
931
+ {
932
+ "epoch": 0.33078880407124683,
933
+ "grad_norm": 2.536672830581665,
934
+ "learning_rate": 2.9754028837998307e-05,
935
+ "loss": 0.0109,
936
+ "step": 1300
937
+ },
938
+ {
939
+ "epoch": 0.3333333333333333,
940
+ "grad_norm": 2.7068772315979004,
941
+ "learning_rate": 2.964093864857224e-05,
942
+ "loss": 0.012,
943
+ "step": 1310
944
+ },
945
+ {
946
+ "epoch": 0.33587786259541985,
947
+ "grad_norm": 0.03697170689702034,
948
+ "learning_rate": 2.9527848459146173e-05,
949
+ "loss": 0.01,
950
+ "step": 1320
951
+ },
952
+ {
953
+ "epoch": 0.3384223918575064,
954
+ "grad_norm": 0.17284952104091644,
955
+ "learning_rate": 2.9414758269720102e-05,
956
+ "loss": 0.003,
957
+ "step": 1330
958
+ },
959
+ {
960
+ "epoch": 0.34096692111959287,
961
+ "grad_norm": 3.8745925426483154,
962
+ "learning_rate": 2.9301668080294038e-05,
963
+ "loss": 0.0156,
964
+ "step": 1340
965
+ },
966
+ {
967
+ "epoch": 0.3435114503816794,
968
+ "grad_norm": 0.44607672095298767,
969
+ "learning_rate": 2.918857789086797e-05,
970
+ "loss": 0.0115,
971
+ "step": 1350
972
+ },
973
+ {
974
+ "epoch": 0.3460559796437659,
975
+ "grad_norm": 6.223470687866211,
976
+ "learning_rate": 2.9075487701441903e-05,
977
+ "loss": 0.0149,
978
+ "step": 1360
979
+ },
980
+ {
981
+ "epoch": 0.3486005089058524,
982
+ "grad_norm": 0.10045173764228821,
983
+ "learning_rate": 2.8962397512015835e-05,
984
+ "loss": 0.0063,
985
+ "step": 1370
986
+ },
987
+ {
988
+ "epoch": 0.3511450381679389,
989
+ "grad_norm": 2.6154091358184814,
990
+ "learning_rate": 2.8849307322589765e-05,
991
+ "loss": 0.0099,
992
+ "step": 1380
993
+ },
994
+ {
995
+ "epoch": 0.35368956743002544,
996
+ "grad_norm": 1.6079035997390747,
997
+ "learning_rate": 2.87362171331637e-05,
998
+ "loss": 0.0079,
999
+ "step": 1390
1000
+ },
1001
+ {
1002
+ "epoch": 0.356234096692112,
1003
+ "grad_norm": 0.3847286105155945,
1004
+ "learning_rate": 2.8623126943737633e-05,
1005
+ "loss": 0.0094,
1006
+ "step": 1400
1007
+ },
1008
+ {
1009
+ "epoch": 0.35877862595419846,
1010
+ "grad_norm": 0.762495219707489,
1011
+ "learning_rate": 2.8510036754311566e-05,
1012
+ "loss": 0.0138,
1013
+ "step": 1410
1014
+ },
1015
+ {
1016
+ "epoch": 0.361323155216285,
1017
+ "grad_norm": 2.905803680419922,
1018
+ "learning_rate": 2.8396946564885498e-05,
1019
+ "loss": 0.0126,
1020
+ "step": 1420
1021
+ },
1022
+ {
1023
+ "epoch": 0.3638676844783715,
1024
+ "grad_norm": 0.20091155171394348,
1025
+ "learning_rate": 2.8283856375459434e-05,
1026
+ "loss": 0.005,
1027
+ "step": 1430
1028
+ },
1029
+ {
1030
+ "epoch": 0.366412213740458,
1031
+ "grad_norm": 0.2232949584722519,
1032
+ "learning_rate": 2.8170766186033363e-05,
1033
+ "loss": 0.0157,
1034
+ "step": 1440
1035
+ },
1036
+ {
1037
+ "epoch": 0.36895674300254455,
1038
+ "grad_norm": 0.11827141046524048,
1039
+ "learning_rate": 2.8057675996607296e-05,
1040
+ "loss": 0.0076,
1041
+ "step": 1450
1042
+ },
1043
+ {
1044
+ "epoch": 0.37150127226463103,
1045
+ "grad_norm": 0.0741715356707573,
1046
+ "learning_rate": 2.794458580718123e-05,
1047
+ "loss": 0.0117,
1048
+ "step": 1460
1049
+ },
1050
+ {
1051
+ "epoch": 0.37404580152671757,
1052
+ "grad_norm": 1.2536979913711548,
1053
+ "learning_rate": 2.783149561775516e-05,
1054
+ "loss": 0.0017,
1055
+ "step": 1470
1056
+ },
1057
+ {
1058
+ "epoch": 0.37659033078880405,
1059
+ "grad_norm": 1.9891741275787354,
1060
+ "learning_rate": 2.7718405428329097e-05,
1061
+ "loss": 0.0109,
1062
+ "step": 1480
1063
+ },
1064
+ {
1065
+ "epoch": 0.3791348600508906,
1066
+ "grad_norm": 1.0660837888717651,
1067
+ "learning_rate": 2.7605315238903026e-05,
1068
+ "loss": 0.007,
1069
+ "step": 1490
1070
+ },
1071
+ {
1072
+ "epoch": 0.3816793893129771,
1073
+ "grad_norm": 0.0798741951584816,
1074
+ "learning_rate": 2.749222504947696e-05,
1075
+ "loss": 0.0139,
1076
+ "step": 1500
1077
+ },
1078
+ {
1079
+ "epoch": 0.3816793893129771,
1080
+ "eval_loss": 0.0349714532494545,
1081
+ "eval_runtime": 100.7459,
1082
+ "eval_samples_per_second": 79.229,
1083
+ "eval_steps_per_second": 0.625,
1084
+ "step": 1500
1085
+ },
1086
+ {
1087
+ "epoch": 0.3842239185750636,
1088
+ "grad_norm": 3.056962013244629,
1089
+ "learning_rate": 2.737913486005089e-05,
1090
+ "loss": 0.019,
1091
+ "step": 1510
1092
+ },
1093
+ {
1094
+ "epoch": 0.38676844783715014,
1095
+ "grad_norm": 0.41486862301826477,
1096
+ "learning_rate": 2.7266044670624827e-05,
1097
+ "loss": 0.0027,
1098
+ "step": 1520
1099
+ },
1100
+ {
1101
+ "epoch": 0.3893129770992366,
1102
+ "grad_norm": 1.0251950025558472,
1103
+ "learning_rate": 2.715295448119876e-05,
1104
+ "loss": 0.0078,
1105
+ "step": 1530
1106
+ },
1107
+ {
1108
+ "epoch": 0.39185750636132316,
1109
+ "grad_norm": 0.019924340769648552,
1110
+ "learning_rate": 2.703986429177269e-05,
1111
+ "loss": 0.0031,
1112
+ "step": 1540
1113
+ },
1114
+ {
1115
+ "epoch": 0.3944020356234097,
1116
+ "grad_norm": 3.1143898963928223,
1117
+ "learning_rate": 2.692677410234662e-05,
1118
+ "loss": 0.0054,
1119
+ "step": 1550
1120
+ },
1121
+ {
1122
+ "epoch": 0.3969465648854962,
1123
+ "grad_norm": 3.4814693927764893,
1124
+ "learning_rate": 2.6813683912920558e-05,
1125
+ "loss": 0.0169,
1126
+ "step": 1560
1127
+ },
1128
+ {
1129
+ "epoch": 0.3994910941475827,
1130
+ "grad_norm": 0.34978416562080383,
1131
+ "learning_rate": 2.670059372349449e-05,
1132
+ "loss": 0.0017,
1133
+ "step": 1570
1134
+ },
1135
+ {
1136
+ "epoch": 0.4020356234096692,
1137
+ "grad_norm": 0.17419882118701935,
1138
+ "learning_rate": 2.6587503534068423e-05,
1139
+ "loss": 0.0017,
1140
+ "step": 1580
1141
+ },
1142
+ {
1143
+ "epoch": 0.40458015267175573,
1144
+ "grad_norm": 6.330338001251221,
1145
+ "learning_rate": 2.6474413344642352e-05,
1146
+ "loss": 0.0081,
1147
+ "step": 1590
1148
+ },
1149
+ {
1150
+ "epoch": 0.4071246819338422,
1151
+ "grad_norm": 0.014109604060649872,
1152
+ "learning_rate": 2.6361323155216285e-05,
1153
+ "loss": 0.0059,
1154
+ "step": 1600
1155
+ },
1156
+ {
1157
+ "epoch": 0.40966921119592875,
1158
+ "grad_norm": 0.06557715684175491,
1159
+ "learning_rate": 2.624823296579022e-05,
1160
+ "loss": 0.0025,
1161
+ "step": 1610
1162
+ },
1163
+ {
1164
+ "epoch": 0.4122137404580153,
1165
+ "grad_norm": 0.06417052447795868,
1166
+ "learning_rate": 2.6135142776364153e-05,
1167
+ "loss": 0.001,
1168
+ "step": 1620
1169
+ },
1170
+ {
1171
+ "epoch": 0.41475826972010177,
1172
+ "grad_norm": 2.6139402389526367,
1173
+ "learning_rate": 2.6022052586938086e-05,
1174
+ "loss": 0.0049,
1175
+ "step": 1630
1176
+ },
1177
+ {
1178
+ "epoch": 0.4173027989821883,
1179
+ "grad_norm": 2.0611274242401123,
1180
+ "learning_rate": 2.5908962397512015e-05,
1181
+ "loss": 0.0062,
1182
+ "step": 1640
1183
+ },
1184
+ {
1185
+ "epoch": 0.4198473282442748,
1186
+ "grad_norm": 0.0023517871741205454,
1187
+ "learning_rate": 2.5795872208085954e-05,
1188
+ "loss": 0.0013,
1189
+ "step": 1650
1190
+ },
1191
+ {
1192
+ "epoch": 0.4223918575063613,
1193
+ "grad_norm": 0.15409255027770996,
1194
+ "learning_rate": 2.5682782018659883e-05,
1195
+ "loss": 0.0131,
1196
+ "step": 1660
1197
+ },
1198
+ {
1199
+ "epoch": 0.42493638676844786,
1200
+ "grad_norm": 0.07905858010053635,
1201
+ "learning_rate": 2.5569691829233816e-05,
1202
+ "loss": 0.0037,
1203
+ "step": 1670
1204
+ },
1205
+ {
1206
+ "epoch": 0.42748091603053434,
1207
+ "grad_norm": 0.3997408151626587,
1208
+ "learning_rate": 2.545660163980775e-05,
1209
+ "loss": 0.0064,
1210
+ "step": 1680
1211
+ },
1212
+ {
1213
+ "epoch": 0.4300254452926209,
1214
+ "grad_norm": 0.14963960647583008,
1215
+ "learning_rate": 2.5343511450381678e-05,
1216
+ "loss": 0.0107,
1217
+ "step": 1690
1218
+ },
1219
+ {
1220
+ "epoch": 0.43256997455470736,
1221
+ "grad_norm": 0.06408097594976425,
1222
+ "learning_rate": 2.5230421260955617e-05,
1223
+ "loss": 0.0072,
1224
+ "step": 1700
1225
+ },
1226
+ {
1227
+ "epoch": 0.4351145038167939,
1228
+ "grad_norm": 3.4469339847564697,
1229
+ "learning_rate": 2.5117331071529546e-05,
1230
+ "loss": 0.008,
1231
+ "step": 1710
1232
+ },
1233
+ {
1234
+ "epoch": 0.43765903307888043,
1235
+ "grad_norm": 0.004595658276230097,
1236
+ "learning_rate": 2.500424088210348e-05,
1237
+ "loss": 0.0081,
1238
+ "step": 1720
1239
+ },
1240
+ {
1241
+ "epoch": 0.4402035623409669,
1242
+ "grad_norm": 4.713712215423584,
1243
+ "learning_rate": 2.489115069267741e-05,
1244
+ "loss": 0.0107,
1245
+ "step": 1730
1246
+ },
1247
+ {
1248
+ "epoch": 0.44274809160305345,
1249
+ "grad_norm": 0.08359724283218384,
1250
+ "learning_rate": 2.4778060503251347e-05,
1251
+ "loss": 0.0018,
1252
+ "step": 1740
1253
+ },
1254
+ {
1255
+ "epoch": 0.44529262086513993,
1256
+ "grad_norm": 2.5329370498657227,
1257
+ "learning_rate": 2.466497031382528e-05,
1258
+ "loss": 0.0086,
1259
+ "step": 1750
1260
+ },
1261
+ {
1262
+ "epoch": 0.44783715012722647,
1263
+ "grad_norm": 0.017571361735463142,
1264
+ "learning_rate": 2.455188012439921e-05,
1265
+ "loss": 0.0065,
1266
+ "step": 1760
1267
+ },
1268
+ {
1269
+ "epoch": 0.45038167938931295,
1270
+ "grad_norm": 0.40600821375846863,
1271
+ "learning_rate": 2.443878993497314e-05,
1272
+ "loss": 0.0062,
1273
+ "step": 1770
1274
+ },
1275
+ {
1276
+ "epoch": 0.4529262086513995,
1277
+ "grad_norm": 3.7990636825561523,
1278
+ "learning_rate": 2.4325699745547078e-05,
1279
+ "loss": 0.0134,
1280
+ "step": 1780
1281
+ },
1282
+ {
1283
+ "epoch": 0.455470737913486,
1284
+ "grad_norm": 0.09965948760509491,
1285
+ "learning_rate": 2.421260955612101e-05,
1286
+ "loss": 0.0068,
1287
+ "step": 1790
1288
+ },
1289
+ {
1290
+ "epoch": 0.4580152671755725,
1291
+ "grad_norm": 2.918208360671997,
1292
+ "learning_rate": 2.4099519366694943e-05,
1293
+ "loss": 0.0093,
1294
+ "step": 1800
1295
+ },
1296
+ {
1297
+ "epoch": 0.46055979643765904,
1298
+ "grad_norm": 0.011430525220930576,
1299
+ "learning_rate": 2.3986429177268872e-05,
1300
+ "loss": 0.0082,
1301
+ "step": 1810
1302
+ },
1303
+ {
1304
+ "epoch": 0.4631043256997455,
1305
+ "grad_norm": 0.15496379137039185,
1306
+ "learning_rate": 2.3873338987842804e-05,
1307
+ "loss": 0.0096,
1308
+ "step": 1820
1309
+ },
1310
+ {
1311
+ "epoch": 0.46564885496183206,
1312
+ "grad_norm": 0.02492368035018444,
1313
+ "learning_rate": 2.376024879841674e-05,
1314
+ "loss": 0.0078,
1315
+ "step": 1830
1316
+ },
1317
+ {
1318
+ "epoch": 0.4681933842239186,
1319
+ "grad_norm": 0.05971188098192215,
1320
+ "learning_rate": 2.3647158608990673e-05,
1321
+ "loss": 0.0089,
1322
+ "step": 1840
1323
+ },
1324
+ {
1325
+ "epoch": 0.4707379134860051,
1326
+ "grad_norm": 0.4808364808559418,
1327
+ "learning_rate": 2.3534068419564605e-05,
1328
+ "loss": 0.0047,
1329
+ "step": 1850
1330
+ },
1331
+ {
1332
+ "epoch": 0.4732824427480916,
1333
+ "grad_norm": 2.379042863845825,
1334
+ "learning_rate": 2.3420978230138535e-05,
1335
+ "loss": 0.0118,
1336
+ "step": 1860
1337
+ },
1338
+ {
1339
+ "epoch": 0.4758269720101781,
1340
+ "grad_norm": 0.024242276325821877,
1341
+ "learning_rate": 2.330788804071247e-05,
1342
+ "loss": 0.009,
1343
+ "step": 1870
1344
+ },
1345
+ {
1346
+ "epoch": 0.47837150127226463,
1347
+ "grad_norm": 1.4494438171386719,
1348
+ "learning_rate": 2.3194797851286403e-05,
1349
+ "loss": 0.0035,
1350
+ "step": 1880
1351
+ },
1352
+ {
1353
+ "epoch": 0.48091603053435117,
1354
+ "grad_norm": 4.915983200073242,
1355
+ "learning_rate": 2.3081707661860336e-05,
1356
+ "loss": 0.0145,
1357
+ "step": 1890
1358
+ },
1359
+ {
1360
+ "epoch": 0.48346055979643765,
1361
+ "grad_norm": 2.4147846698760986,
1362
+ "learning_rate": 2.296861747243427e-05,
1363
+ "loss": 0.0078,
1364
+ "step": 1900
1365
+ },
1366
+ {
1367
+ "epoch": 0.4860050890585242,
1368
+ "grad_norm": 0.41452011466026306,
1369
+ "learning_rate": 2.2855527283008204e-05,
1370
+ "loss": 0.0036,
1371
+ "step": 1910
1372
+ },
1373
+ {
1374
+ "epoch": 0.48854961832061067,
1375
+ "grad_norm": 0.6090364456176758,
1376
+ "learning_rate": 2.2742437093582133e-05,
1377
+ "loss": 0.0054,
1378
+ "step": 1920
1379
+ },
1380
+ {
1381
+ "epoch": 0.4910941475826972,
1382
+ "grad_norm": 2.908536434173584,
1383
+ "learning_rate": 2.2629346904156066e-05,
1384
+ "loss": 0.0055,
1385
+ "step": 1930
1386
+ },
1387
+ {
1388
+ "epoch": 0.49363867684478374,
1389
+ "grad_norm": 3.4067792892456055,
1390
+ "learning_rate": 2.251625671473e-05,
1391
+ "loss": 0.0063,
1392
+ "step": 1940
1393
+ },
1394
+ {
1395
+ "epoch": 0.4961832061068702,
1396
+ "grad_norm": 0.12075275182723999,
1397
+ "learning_rate": 2.240316652530393e-05,
1398
+ "loss": 0.0075,
1399
+ "step": 1950
1400
+ },
1401
+ {
1402
+ "epoch": 0.49872773536895676,
1403
+ "grad_norm": 0.10985302925109863,
1404
+ "learning_rate": 2.2290076335877867e-05,
1405
+ "loss": 0.0032,
1406
+ "step": 1960
1407
+ },
1408
+ {
1409
+ "epoch": 0.5012722646310432,
1410
+ "grad_norm": 1.9272608757019043,
1411
+ "learning_rate": 2.2176986146451796e-05,
1412
+ "loss": 0.0069,
1413
+ "step": 1970
1414
+ },
1415
+ {
1416
+ "epoch": 0.5038167938931297,
1417
+ "grad_norm": 3.202061891555786,
1418
+ "learning_rate": 2.206389595702573e-05,
1419
+ "loss": 0.0073,
1420
+ "step": 1980
1421
+ },
1422
+ {
1423
+ "epoch": 0.5063613231552163,
1424
+ "grad_norm": 1.0483746528625488,
1425
+ "learning_rate": 2.195080576759966e-05,
1426
+ "loss": 0.0031,
1427
+ "step": 1990
1428
+ },
1429
+ {
1430
+ "epoch": 0.5089058524173028,
1431
+ "grad_norm": 1.31486976146698,
1432
+ "learning_rate": 2.1837715578173597e-05,
1433
+ "loss": 0.0025,
1434
+ "step": 2000
1435
+ },
1436
+ {
1437
+ "epoch": 0.5089058524173028,
1438
+ "eval_loss": 0.034609079360961914,
1439
+ "eval_runtime": 100.7933,
1440
+ "eval_samples_per_second": 79.192,
1441
+ "eval_steps_per_second": 0.625,
1442
+ "step": 2000
1443
+ },
1444
+ {
1445
+ "epoch": 0.5114503816793893,
1446
+ "grad_norm": 0.05615166202187538,
1447
+ "learning_rate": 2.172462538874753e-05,
1448
+ "loss": 0.0114,
1449
+ "step": 2010
1450
+ },
1451
+ {
1452
+ "epoch": 0.5139949109414759,
1453
+ "grad_norm": 1.326418161392212,
1454
+ "learning_rate": 2.161153519932146e-05,
1455
+ "loss": 0.0053,
1456
+ "step": 2020
1457
+ },
1458
+ {
1459
+ "epoch": 0.5165394402035624,
1460
+ "grad_norm": 1.4809695482254028,
1461
+ "learning_rate": 2.1498445009895392e-05,
1462
+ "loss": 0.0085,
1463
+ "step": 2030
1464
+ },
1465
+ {
1466
+ "epoch": 0.5190839694656488,
1467
+ "grad_norm": 2.975813388824463,
1468
+ "learning_rate": 2.1385354820469324e-05,
1469
+ "loss": 0.0101,
1470
+ "step": 2040
1471
+ },
1472
+ {
1473
+ "epoch": 0.5216284987277354,
1474
+ "grad_norm": 0.8962512612342834,
1475
+ "learning_rate": 2.127226463104326e-05,
1476
+ "loss": 0.0104,
1477
+ "step": 2050
1478
+ },
1479
+ {
1480
+ "epoch": 0.5241730279898219,
1481
+ "grad_norm": 3.6851797103881836,
1482
+ "learning_rate": 2.1159174441617193e-05,
1483
+ "loss": 0.0074,
1484
+ "step": 2060
1485
+ },
1486
+ {
1487
+ "epoch": 0.5267175572519084,
1488
+ "grad_norm": 2.4362735748291016,
1489
+ "learning_rate": 2.1046084252191122e-05,
1490
+ "loss": 0.0119,
1491
+ "step": 2070
1492
+ },
1493
+ {
1494
+ "epoch": 0.5292620865139949,
1495
+ "grad_norm": 3.4005730152130127,
1496
+ "learning_rate": 2.0932994062765055e-05,
1497
+ "loss": 0.0179,
1498
+ "step": 2080
1499
+ },
1500
+ {
1501
+ "epoch": 0.5318066157760815,
1502
+ "grad_norm": 0.020965535193681717,
1503
+ "learning_rate": 2.081990387333899e-05,
1504
+ "loss": 0.0108,
1505
+ "step": 2090
1506
+ },
1507
+ {
1508
+ "epoch": 0.5343511450381679,
1509
+ "grad_norm": 3.1642138957977295,
1510
+ "learning_rate": 2.0706813683912923e-05,
1511
+ "loss": 0.0089,
1512
+ "step": 2100
1513
+ },
1514
+ {
1515
+ "epoch": 0.5368956743002544,
1516
+ "grad_norm": 1.4003890752792358,
1517
+ "learning_rate": 2.0593723494486856e-05,
1518
+ "loss": 0.0086,
1519
+ "step": 2110
1520
+ },
1521
+ {
1522
+ "epoch": 0.539440203562341,
1523
+ "grad_norm": 3.2447030544281006,
1524
+ "learning_rate": 2.0480633305060785e-05,
1525
+ "loss": 0.0092,
1526
+ "step": 2120
1527
+ },
1528
+ {
1529
+ "epoch": 0.5419847328244275,
1530
+ "grad_norm": 0.20196814835071564,
1531
+ "learning_rate": 2.0367543115634724e-05,
1532
+ "loss": 0.0091,
1533
+ "step": 2130
1534
+ },
1535
+ {
1536
+ "epoch": 0.544529262086514,
1537
+ "grad_norm": 0.09086956083774567,
1538
+ "learning_rate": 2.0254452926208653e-05,
1539
+ "loss": 0.0115,
1540
+ "step": 2140
1541
+ },
1542
+ {
1543
+ "epoch": 0.5470737913486005,
1544
+ "grad_norm": 0.8166348934173584,
1545
+ "learning_rate": 2.0141362736782586e-05,
1546
+ "loss": 0.0138,
1547
+ "step": 2150
1548
+ },
1549
+ {
1550
+ "epoch": 0.549618320610687,
1551
+ "grad_norm": 1.8422399759292603,
1552
+ "learning_rate": 2.002827254735652e-05,
1553
+ "loss": 0.0041,
1554
+ "step": 2160
1555
+ },
1556
+ {
1557
+ "epoch": 0.5521628498727735,
1558
+ "grad_norm": 1.3676835298538208,
1559
+ "learning_rate": 1.991518235793045e-05,
1560
+ "loss": 0.0051,
1561
+ "step": 2170
1562
+ },
1563
+ {
1564
+ "epoch": 0.55470737913486,
1565
+ "grad_norm": 3.4715235233306885,
1566
+ "learning_rate": 1.9802092168504384e-05,
1567
+ "loss": 0.0101,
1568
+ "step": 2180
1569
+ },
1570
+ {
1571
+ "epoch": 0.5572519083969466,
1572
+ "grad_norm": 0.4825442433357239,
1573
+ "learning_rate": 1.9689001979078316e-05,
1574
+ "loss": 0.0065,
1575
+ "step": 2190
1576
+ },
1577
+ {
1578
+ "epoch": 0.5597964376590331,
1579
+ "grad_norm": 0.03331977128982544,
1580
+ "learning_rate": 1.957591178965225e-05,
1581
+ "loss": 0.0136,
1582
+ "step": 2200
1583
+ },
1584
+ {
1585
+ "epoch": 0.5623409669211196,
1586
+ "grad_norm": 0.6727198362350464,
1587
+ "learning_rate": 1.946282160022618e-05,
1588
+ "loss": 0.0023,
1589
+ "step": 2210
1590
+ },
1591
+ {
1592
+ "epoch": 0.5648854961832062,
1593
+ "grad_norm": 2.1888349056243896,
1594
+ "learning_rate": 1.9349731410800114e-05,
1595
+ "loss": 0.0095,
1596
+ "step": 2220
1597
+ },
1598
+ {
1599
+ "epoch": 0.5674300254452926,
1600
+ "grad_norm": 4.135168075561523,
1601
+ "learning_rate": 1.923664122137405e-05,
1602
+ "loss": 0.0179,
1603
+ "step": 2230
1604
+ },
1605
+ {
1606
+ "epoch": 0.5699745547073791,
1607
+ "grad_norm": 1.3800098896026611,
1608
+ "learning_rate": 1.912355103194798e-05,
1609
+ "loss": 0.0045,
1610
+ "step": 2240
1611
+ },
1612
+ {
1613
+ "epoch": 0.5725190839694656,
1614
+ "grad_norm": 0.2383573353290558,
1615
+ "learning_rate": 1.901046084252191e-05,
1616
+ "loss": 0.0154,
1617
+ "step": 2250
1618
+ },
1619
+ {
1620
+ "epoch": 0.5750636132315522,
1621
+ "grad_norm": 0.2643699645996094,
1622
+ "learning_rate": 1.8897370653095844e-05,
1623
+ "loss": 0.0065,
1624
+ "step": 2260
1625
+ },
1626
+ {
1627
+ "epoch": 0.5776081424936387,
1628
+ "grad_norm": 0.7938392162322998,
1629
+ "learning_rate": 1.8784280463669777e-05,
1630
+ "loss": 0.0074,
1631
+ "step": 2270
1632
+ },
1633
+ {
1634
+ "epoch": 0.5801526717557252,
1635
+ "grad_norm": 1.8504903316497803,
1636
+ "learning_rate": 1.8671190274243713e-05,
1637
+ "loss": 0.0093,
1638
+ "step": 2280
1639
+ },
1640
+ {
1641
+ "epoch": 0.5826972010178118,
1642
+ "grad_norm": 3.9438815116882324,
1643
+ "learning_rate": 1.8558100084817642e-05,
1644
+ "loss": 0.0065,
1645
+ "step": 2290
1646
+ },
1647
+ {
1648
+ "epoch": 0.5852417302798982,
1649
+ "grad_norm": 0.024695103988051414,
1650
+ "learning_rate": 1.8445009895391578e-05,
1651
+ "loss": 0.0007,
1652
+ "step": 2300
1653
+ },
1654
+ {
1655
+ "epoch": 0.5877862595419847,
1656
+ "grad_norm": 0.5728102326393127,
1657
+ "learning_rate": 1.8331919705965507e-05,
1658
+ "loss": 0.0014,
1659
+ "step": 2310
1660
+ },
1661
+ {
1662
+ "epoch": 0.5903307888040712,
1663
+ "grad_norm": 0.0501931831240654,
1664
+ "learning_rate": 1.8218829516539443e-05,
1665
+ "loss": 0.006,
1666
+ "step": 2320
1667
+ },
1668
+ {
1669
+ "epoch": 0.5928753180661578,
1670
+ "grad_norm": 1.13953697681427,
1671
+ "learning_rate": 1.8105739327113376e-05,
1672
+ "loss": 0.006,
1673
+ "step": 2330
1674
+ },
1675
+ {
1676
+ "epoch": 0.5954198473282443,
1677
+ "grad_norm": 2.1232895851135254,
1678
+ "learning_rate": 1.7992649137687308e-05,
1679
+ "loss": 0.0073,
1680
+ "step": 2340
1681
+ },
1682
+ {
1683
+ "epoch": 0.5979643765903307,
1684
+ "grad_norm": 0.37371209263801575,
1685
+ "learning_rate": 1.787955894826124e-05,
1686
+ "loss": 0.008,
1687
+ "step": 2350
1688
+ },
1689
+ {
1690
+ "epoch": 0.6005089058524173,
1691
+ "grad_norm": 2.0648906230926514,
1692
+ "learning_rate": 1.776646875883517e-05,
1693
+ "loss": 0.0063,
1694
+ "step": 2360
1695
+ },
1696
+ {
1697
+ "epoch": 0.6030534351145038,
1698
+ "grad_norm": 2.0410525798797607,
1699
+ "learning_rate": 1.7653378569409106e-05,
1700
+ "loss": 0.0035,
1701
+ "step": 2370
1702
+ },
1703
+ {
1704
+ "epoch": 0.6055979643765903,
1705
+ "grad_norm": 0.1869073212146759,
1706
+ "learning_rate": 1.754028837998304e-05,
1707
+ "loss": 0.0088,
1708
+ "step": 2380
1709
+ },
1710
+ {
1711
+ "epoch": 0.6081424936386769,
1712
+ "grad_norm": 2.5563478469848633,
1713
+ "learning_rate": 1.742719819055697e-05,
1714
+ "loss": 0.0092,
1715
+ "step": 2390
1716
+ },
1717
+ {
1718
+ "epoch": 0.6106870229007634,
1719
+ "grad_norm": 2.4307727813720703,
1720
+ "learning_rate": 1.7314108001130904e-05,
1721
+ "loss": 0.0101,
1722
+ "step": 2400
1723
+ },
1724
+ {
1725
+ "epoch": 0.6132315521628499,
1726
+ "grad_norm": 3.2351479530334473,
1727
+ "learning_rate": 1.7201017811704836e-05,
1728
+ "loss": 0.0104,
1729
+ "step": 2410
1730
+ },
1731
+ {
1732
+ "epoch": 0.6157760814249363,
1733
+ "grad_norm": 3.1805145740509033,
1734
+ "learning_rate": 1.708792762227877e-05,
1735
+ "loss": 0.0129,
1736
+ "step": 2420
1737
+ },
1738
+ {
1739
+ "epoch": 0.6183206106870229,
1740
+ "grad_norm": 1.25412118434906,
1741
+ "learning_rate": 1.69748374328527e-05,
1742
+ "loss": 0.0024,
1743
+ "step": 2430
1744
+ },
1745
+ {
1746
+ "epoch": 0.6208651399491094,
1747
+ "grad_norm": 0.02675499953329563,
1748
+ "learning_rate": 1.6861747243426634e-05,
1749
+ "loss": 0.0066,
1750
+ "step": 2440
1751
+ },
1752
+ {
1753
+ "epoch": 0.6234096692111959,
1754
+ "grad_norm": 5.5407819747924805,
1755
+ "learning_rate": 1.6748657054000566e-05,
1756
+ "loss": 0.007,
1757
+ "step": 2450
1758
+ },
1759
+ {
1760
+ "epoch": 0.6259541984732825,
1761
+ "grad_norm": 3.5257675647735596,
1762
+ "learning_rate": 1.66355668645745e-05,
1763
+ "loss": 0.0139,
1764
+ "step": 2460
1765
+ },
1766
+ {
1767
+ "epoch": 0.628498727735369,
1768
+ "grad_norm": 1.163690209388733,
1769
+ "learning_rate": 1.652247667514843e-05,
1770
+ "loss": 0.0041,
1771
+ "step": 2470
1772
+ },
1773
+ {
1774
+ "epoch": 0.6310432569974554,
1775
+ "grad_norm": 1.5052272081375122,
1776
+ "learning_rate": 1.6409386485722364e-05,
1777
+ "loss": 0.0162,
1778
+ "step": 2480
1779
+ },
1780
+ {
1781
+ "epoch": 0.6335877862595419,
1782
+ "grad_norm": 0.1098732277750969,
1783
+ "learning_rate": 1.6296296296296297e-05,
1784
+ "loss": 0.0077,
1785
+ "step": 2490
1786
+ },
1787
+ {
1788
+ "epoch": 0.6361323155216285,
1789
+ "grad_norm": 2.6149792671203613,
1790
+ "learning_rate": 1.618320610687023e-05,
1791
+ "loss": 0.0042,
1792
+ "step": 2500
1793
+ },
1794
+ {
1795
+ "epoch": 0.6361323155216285,
1796
+ "eval_loss": 0.030483221635222435,
1797
+ "eval_runtime": 100.851,
1798
+ "eval_samples_per_second": 79.146,
1799
+ "eval_steps_per_second": 0.625,
1800
+ "step": 2500
1801
+ },
1802
+ {
1803
+ "epoch": 0.638676844783715,
1804
+ "grad_norm": 0.5812219977378845,
1805
+ "learning_rate": 1.6070115917444162e-05,
1806
+ "loss": 0.0107,
1807
+ "step": 2510
1808
+ },
1809
+ {
1810
+ "epoch": 0.6412213740458015,
1811
+ "grad_norm": 2.341726303100586,
1812
+ "learning_rate": 1.5957025728018098e-05,
1813
+ "loss": 0.0097,
1814
+ "step": 2520
1815
+ },
1816
+ {
1817
+ "epoch": 0.6437659033078881,
1818
+ "grad_norm": 1.1993039846420288,
1819
+ "learning_rate": 1.5843935538592027e-05,
1820
+ "loss": 0.0088,
1821
+ "step": 2530
1822
+ },
1823
+ {
1824
+ "epoch": 0.6463104325699746,
1825
+ "grad_norm": 0.019058074802160263,
1826
+ "learning_rate": 1.5730845349165963e-05,
1827
+ "loss": 0.0049,
1828
+ "step": 2540
1829
+ },
1830
+ {
1831
+ "epoch": 0.648854961832061,
1832
+ "grad_norm": 1.7715319395065308,
1833
+ "learning_rate": 1.5617755159739892e-05,
1834
+ "loss": 0.0077,
1835
+ "step": 2550
1836
+ },
1837
+ {
1838
+ "epoch": 0.6513994910941476,
1839
+ "grad_norm": 2.300363302230835,
1840
+ "learning_rate": 1.5504664970313828e-05,
1841
+ "loss": 0.0289,
1842
+ "step": 2560
1843
+ },
1844
+ {
1845
+ "epoch": 0.6539440203562341,
1846
+ "grad_norm": 3.3478307723999023,
1847
+ "learning_rate": 1.539157478088776e-05,
1848
+ "loss": 0.0101,
1849
+ "step": 2570
1850
+ },
1851
+ {
1852
+ "epoch": 0.6564885496183206,
1853
+ "grad_norm": 0.5511828660964966,
1854
+ "learning_rate": 1.5278484591461693e-05,
1855
+ "loss": 0.013,
1856
+ "step": 2580
1857
+ },
1858
+ {
1859
+ "epoch": 0.6590330788804071,
1860
+ "grad_norm": 3.244690418243408,
1861
+ "learning_rate": 1.5165394402035624e-05,
1862
+ "loss": 0.0129,
1863
+ "step": 2590
1864
+ },
1865
+ {
1866
+ "epoch": 0.6615776081424937,
1867
+ "grad_norm": 0.08320210874080658,
1868
+ "learning_rate": 1.5052304212609557e-05,
1869
+ "loss": 0.004,
1870
+ "step": 2600
1871
+ },
1872
+ {
1873
+ "epoch": 0.6641221374045801,
1874
+ "grad_norm": 0.06972193717956543,
1875
+ "learning_rate": 1.493921402318349e-05,
1876
+ "loss": 0.0077,
1877
+ "step": 2610
1878
+ },
1879
+ {
1880
+ "epoch": 0.6666666666666666,
1881
+ "grad_norm": 0.010952036827802658,
1882
+ "learning_rate": 1.4826123833757422e-05,
1883
+ "loss": 0.0045,
1884
+ "step": 2620
1885
+ },
1886
+ {
1887
+ "epoch": 0.6692111959287532,
1888
+ "grad_norm": 0.1785265952348709,
1889
+ "learning_rate": 1.4713033644331356e-05,
1890
+ "loss": 0.0011,
1891
+ "step": 2630
1892
+ },
1893
+ {
1894
+ "epoch": 0.6717557251908397,
1895
+ "grad_norm": 3.3081789016723633,
1896
+ "learning_rate": 1.4599943454905287e-05,
1897
+ "loss": 0.0092,
1898
+ "step": 2640
1899
+ },
1900
+ {
1901
+ "epoch": 0.6743002544529262,
1902
+ "grad_norm": 0.011730634607374668,
1903
+ "learning_rate": 1.4486853265479221e-05,
1904
+ "loss": 0.0019,
1905
+ "step": 2650
1906
+ },
1907
+ {
1908
+ "epoch": 0.6768447837150128,
1909
+ "grad_norm": 1.933732271194458,
1910
+ "learning_rate": 1.4373763076053154e-05,
1911
+ "loss": 0.0026,
1912
+ "step": 2660
1913
+ },
1914
+ {
1915
+ "epoch": 0.6793893129770993,
1916
+ "grad_norm": 2.0810701847076416,
1917
+ "learning_rate": 1.4260672886627088e-05,
1918
+ "loss": 0.0104,
1919
+ "step": 2670
1920
+ },
1921
+ {
1922
+ "epoch": 0.6819338422391857,
1923
+ "grad_norm": 4.2147369384765625,
1924
+ "learning_rate": 1.4147582697201019e-05,
1925
+ "loss": 0.0257,
1926
+ "step": 2680
1927
+ },
1928
+ {
1929
+ "epoch": 0.6844783715012722,
1930
+ "grad_norm": 6.2209014892578125,
1931
+ "learning_rate": 1.4034492507774953e-05,
1932
+ "loss": 0.0112,
1933
+ "step": 2690
1934
+ },
1935
+ {
1936
+ "epoch": 0.6870229007633588,
1937
+ "grad_norm": 4.587314128875732,
1938
+ "learning_rate": 1.3921402318348884e-05,
1939
+ "loss": 0.007,
1940
+ "step": 2700
1941
+ },
1942
+ {
1943
+ "epoch": 0.6895674300254453,
1944
+ "grad_norm": 5.7180256843566895,
1945
+ "learning_rate": 1.3808312128922817e-05,
1946
+ "loss": 0.0107,
1947
+ "step": 2710
1948
+ },
1949
+ {
1950
+ "epoch": 0.6921119592875318,
1951
+ "grad_norm": 1.0686004161834717,
1952
+ "learning_rate": 1.369522193949675e-05,
1953
+ "loss": 0.0136,
1954
+ "step": 2720
1955
+ },
1956
+ {
1957
+ "epoch": 0.6946564885496184,
1958
+ "grad_norm": 3.1578478813171387,
1959
+ "learning_rate": 1.3582131750070682e-05,
1960
+ "loss": 0.0101,
1961
+ "step": 2730
1962
+ },
1963
+ {
1964
+ "epoch": 0.6972010178117048,
1965
+ "grad_norm": 1.4626826047897339,
1966
+ "learning_rate": 1.3469041560644616e-05,
1967
+ "loss": 0.0051,
1968
+ "step": 2740
1969
+ },
1970
+ {
1971
+ "epoch": 0.6997455470737913,
1972
+ "grad_norm": 1.409559726715088,
1973
+ "learning_rate": 1.3355951371218547e-05,
1974
+ "loss": 0.0048,
1975
+ "step": 2750
1976
+ },
1977
+ {
1978
+ "epoch": 0.7022900763358778,
1979
+ "grad_norm": 1.5923247337341309,
1980
+ "learning_rate": 1.3242861181792481e-05,
1981
+ "loss": 0.0052,
1982
+ "step": 2760
1983
+ },
1984
+ {
1985
+ "epoch": 0.7048346055979644,
1986
+ "grad_norm": 0.01753048412501812,
1987
+ "learning_rate": 1.3129770992366414e-05,
1988
+ "loss": 0.0051,
1989
+ "step": 2770
1990
+ },
1991
+ {
1992
+ "epoch": 0.7073791348600509,
1993
+ "grad_norm": 0.022448958829045296,
1994
+ "learning_rate": 1.3016680802940346e-05,
1995
+ "loss": 0.0008,
1996
+ "step": 2780
1997
+ },
1998
+ {
1999
+ "epoch": 0.7099236641221374,
2000
+ "grad_norm": 0.010822678916156292,
2001
+ "learning_rate": 1.2903590613514279e-05,
2002
+ "loss": 0.003,
2003
+ "step": 2790
2004
+ },
2005
+ {
2006
+ "epoch": 0.712468193384224,
2007
+ "grad_norm": 1.4215025901794434,
2008
+ "learning_rate": 1.2790500424088213e-05,
2009
+ "loss": 0.0061,
2010
+ "step": 2800
2011
+ },
2012
+ {
2013
+ "epoch": 0.7150127226463104,
2014
+ "grad_norm": 0.11338794231414795,
2015
+ "learning_rate": 1.2677410234662144e-05,
2016
+ "loss": 0.0045,
2017
+ "step": 2810
2018
+ },
2019
+ {
2020
+ "epoch": 0.7175572519083969,
2021
+ "grad_norm": 2.464766263961792,
2022
+ "learning_rate": 1.2564320045236076e-05,
2023
+ "loss": 0.0045,
2024
+ "step": 2820
2025
+ },
2026
+ {
2027
+ "epoch": 0.7201017811704835,
2028
+ "grad_norm": 0.09698779135942459,
2029
+ "learning_rate": 1.2451229855810009e-05,
2030
+ "loss": 0.0037,
2031
+ "step": 2830
2032
+ },
2033
+ {
2034
+ "epoch": 0.72264631043257,
2035
+ "grad_norm": 0.04683073237538338,
2036
+ "learning_rate": 1.2338139666383942e-05,
2037
+ "loss": 0.0174,
2038
+ "step": 2840
2039
+ },
2040
+ {
2041
+ "epoch": 0.7251908396946565,
2042
+ "grad_norm": 0.013533544726669788,
2043
+ "learning_rate": 1.2225049476957876e-05,
2044
+ "loss": 0.0051,
2045
+ "step": 2850
2046
+ },
2047
+ {
2048
+ "epoch": 0.727735368956743,
2049
+ "grad_norm": 1.7438645362854004,
2050
+ "learning_rate": 1.2111959287531807e-05,
2051
+ "loss": 0.0161,
2052
+ "step": 2860
2053
+ },
2054
+ {
2055
+ "epoch": 0.7302798982188295,
2056
+ "grad_norm": 1.247225046157837,
2057
+ "learning_rate": 1.1998869098105741e-05,
2058
+ "loss": 0.0006,
2059
+ "step": 2870
2060
+ },
2061
+ {
2062
+ "epoch": 0.732824427480916,
2063
+ "grad_norm": 5.181498050689697,
2064
+ "learning_rate": 1.1885778908679672e-05,
2065
+ "loss": 0.0059,
2066
+ "step": 2880
2067
+ },
2068
+ {
2069
+ "epoch": 0.7353689567430025,
2070
+ "grad_norm": 2.543138027191162,
2071
+ "learning_rate": 1.1772688719253606e-05,
2072
+ "loss": 0.015,
2073
+ "step": 2890
2074
+ },
2075
+ {
2076
+ "epoch": 0.7379134860050891,
2077
+ "grad_norm": 0.5215864777565002,
2078
+ "learning_rate": 1.1659598529827539e-05,
2079
+ "loss": 0.0144,
2080
+ "step": 2900
2081
+ },
2082
+ {
2083
+ "epoch": 0.7404580152671756,
2084
+ "grad_norm": 0.017565440386533737,
2085
+ "learning_rate": 1.1546508340401473e-05,
2086
+ "loss": 0.0213,
2087
+ "step": 2910
2088
+ },
2089
+ {
2090
+ "epoch": 0.7430025445292621,
2091
+ "grad_norm": 0.2629789113998413,
2092
+ "learning_rate": 1.1433418150975404e-05,
2093
+ "loss": 0.0051,
2094
+ "step": 2920
2095
+ },
2096
+ {
2097
+ "epoch": 0.7455470737913485,
2098
+ "grad_norm": 0.07063543051481247,
2099
+ "learning_rate": 1.1320327961549336e-05,
2100
+ "loss": 0.0024,
2101
+ "step": 2930
2102
+ },
2103
+ {
2104
+ "epoch": 0.7480916030534351,
2105
+ "grad_norm": 4.683466911315918,
2106
+ "learning_rate": 1.1207237772123269e-05,
2107
+ "loss": 0.0079,
2108
+ "step": 2940
2109
+ },
2110
+ {
2111
+ "epoch": 0.7506361323155216,
2112
+ "grad_norm": 1.2442190647125244,
2113
+ "learning_rate": 1.1094147582697202e-05,
2114
+ "loss": 0.0129,
2115
+ "step": 2950
2116
+ },
2117
+ {
2118
+ "epoch": 0.7531806615776081,
2119
+ "grad_norm": 1.3911083936691284,
2120
+ "learning_rate": 1.0981057393271136e-05,
2121
+ "loss": 0.0057,
2122
+ "step": 2960
2123
+ },
2124
+ {
2125
+ "epoch": 0.7557251908396947,
2126
+ "grad_norm": 2.5614712238311768,
2127
+ "learning_rate": 1.0867967203845067e-05,
2128
+ "loss": 0.0069,
2129
+ "step": 2970
2130
+ },
2131
+ {
2132
+ "epoch": 0.7582697201017812,
2133
+ "grad_norm": 0.22300787270069122,
2134
+ "learning_rate": 1.0754877014419001e-05,
2135
+ "loss": 0.0075,
2136
+ "step": 2980
2137
+ },
2138
+ {
2139
+ "epoch": 0.7608142493638677,
2140
+ "grad_norm": 0.11393741518259048,
2141
+ "learning_rate": 1.0641786824992932e-05,
2142
+ "loss": 0.0042,
2143
+ "step": 2990
2144
+ },
2145
+ {
2146
+ "epoch": 0.7633587786259542,
2147
+ "grad_norm": 3.6102774143218994,
2148
+ "learning_rate": 1.0528696635566866e-05,
2149
+ "loss": 0.0065,
2150
+ "step": 3000
2151
+ },
2152
+ {
2153
+ "epoch": 0.7633587786259542,
2154
+ "eval_loss": 0.028713775798678398,
2155
+ "eval_runtime": 100.7408,
2156
+ "eval_samples_per_second": 79.233,
2157
+ "eval_steps_per_second": 0.625,
2158
+ "step": 3000
2159
+ },
2160
+ {
2161
+ "epoch": 0.7659033078880407,
2162
+ "grad_norm": 0.09737461805343628,
2163
+ "learning_rate": 1.0415606446140799e-05,
2164
+ "loss": 0.0064,
2165
+ "step": 3010
2166
+ },
2167
+ {
2168
+ "epoch": 0.7684478371501272,
2169
+ "grad_norm": 0.17835134267807007,
2170
+ "learning_rate": 1.0302516256714731e-05,
2171
+ "loss": 0.0043,
2172
+ "step": 3020
2173
+ },
2174
+ {
2175
+ "epoch": 0.7709923664122137,
2176
+ "grad_norm": 7.2621588706970215,
2177
+ "learning_rate": 1.0189426067288664e-05,
2178
+ "loss": 0.0051,
2179
+ "step": 3030
2180
+ },
2181
+ {
2182
+ "epoch": 0.7735368956743003,
2183
+ "grad_norm": 1.3994181156158447,
2184
+ "learning_rate": 1.0076335877862595e-05,
2185
+ "loss": 0.0052,
2186
+ "step": 3040
2187
+ },
2188
+ {
2189
+ "epoch": 0.7760814249363868,
2190
+ "grad_norm": 0.3551107943058014,
2191
+ "learning_rate": 9.963245688436529e-06,
2192
+ "loss": 0.0028,
2193
+ "step": 3050
2194
+ },
2195
+ {
2196
+ "epoch": 0.7786259541984732,
2197
+ "grad_norm": 3.8419830799102783,
2198
+ "learning_rate": 9.850155499010461e-06,
2199
+ "loss": 0.0092,
2200
+ "step": 3060
2201
+ },
2202
+ {
2203
+ "epoch": 0.7811704834605598,
2204
+ "grad_norm": 0.03152012452483177,
2205
+ "learning_rate": 9.737065309584394e-06,
2206
+ "loss": 0.0166,
2207
+ "step": 3070
2208
+ },
2209
+ {
2210
+ "epoch": 0.7837150127226463,
2211
+ "grad_norm": 1.2947672605514526,
2212
+ "learning_rate": 9.623975120158328e-06,
2213
+ "loss": 0.0044,
2214
+ "step": 3080
2215
+ },
2216
+ {
2217
+ "epoch": 0.7862595419847328,
2218
+ "grad_norm": 0.0874599888920784,
2219
+ "learning_rate": 9.51088493073226e-06,
2220
+ "loss": 0.0054,
2221
+ "step": 3090
2222
+ },
2223
+ {
2224
+ "epoch": 0.7888040712468194,
2225
+ "grad_norm": 0.06683657318353653,
2226
+ "learning_rate": 9.397794741306192e-06,
2227
+ "loss": 0.0064,
2228
+ "step": 3100
2229
+ },
2230
+ {
2231
+ "epoch": 0.7913486005089059,
2232
+ "grad_norm": 0.09484909474849701,
2233
+ "learning_rate": 9.284704551880124e-06,
2234
+ "loss": 0.0036,
2235
+ "step": 3110
2236
+ },
2237
+ {
2238
+ "epoch": 0.7938931297709924,
2239
+ "grad_norm": 0.599434494972229,
2240
+ "learning_rate": 9.171614362454057e-06,
2241
+ "loss": 0.004,
2242
+ "step": 3120
2243
+ },
2244
+ {
2245
+ "epoch": 0.7964376590330788,
2246
+ "grad_norm": 0.015714434906840324,
2247
+ "learning_rate": 9.058524173027991e-06,
2248
+ "loss": 0.0114,
2249
+ "step": 3130
2250
+ },
2251
+ {
2252
+ "epoch": 0.7989821882951654,
2253
+ "grad_norm": 0.23739086091518402,
2254
+ "learning_rate": 8.945433983601924e-06,
2255
+ "loss": 0.0048,
2256
+ "step": 3140
2257
+ },
2258
+ {
2259
+ "epoch": 0.8015267175572519,
2260
+ "grad_norm": 0.10904980450868607,
2261
+ "learning_rate": 8.832343794175856e-06,
2262
+ "loss": 0.0052,
2263
+ "step": 3150
2264
+ },
2265
+ {
2266
+ "epoch": 0.8040712468193384,
2267
+ "grad_norm": 3.4536166191101074,
2268
+ "learning_rate": 8.719253604749789e-06,
2269
+ "loss": 0.0165,
2270
+ "step": 3160
2271
+ },
2272
+ {
2273
+ "epoch": 0.806615776081425,
2274
+ "grad_norm": 0.13291485607624054,
2275
+ "learning_rate": 8.606163415323721e-06,
2276
+ "loss": 0.0119,
2277
+ "step": 3170
2278
+ },
2279
+ {
2280
+ "epoch": 0.8091603053435115,
2281
+ "grad_norm": 1.6434555053710938,
2282
+ "learning_rate": 8.493073225897654e-06,
2283
+ "loss": 0.009,
2284
+ "step": 3180
2285
+ },
2286
+ {
2287
+ "epoch": 0.811704834605598,
2288
+ "grad_norm": 0.05683699622750282,
2289
+ "learning_rate": 8.379983036471587e-06,
2290
+ "loss": 0.0131,
2291
+ "step": 3190
2292
+ },
2293
+ {
2294
+ "epoch": 0.8142493638676844,
2295
+ "grad_norm": 3.6685760021209717,
2296
+ "learning_rate": 8.266892847045519e-06,
2297
+ "loss": 0.0277,
2298
+ "step": 3200
2299
+ },
2300
+ {
2301
+ "epoch": 0.816793893129771,
2302
+ "grad_norm": 0.07328160107135773,
2303
+ "learning_rate": 8.153802657619452e-06,
2304
+ "loss": 0.002,
2305
+ "step": 3210
2306
+ },
2307
+ {
2308
+ "epoch": 0.8193384223918575,
2309
+ "grad_norm": 4.1098480224609375,
2310
+ "learning_rate": 8.040712468193384e-06,
2311
+ "loss": 0.0046,
2312
+ "step": 3220
2313
+ },
2314
+ {
2315
+ "epoch": 0.821882951653944,
2316
+ "grad_norm": 0.4604232907295227,
2317
+ "learning_rate": 7.927622278767317e-06,
2318
+ "loss": 0.0146,
2319
+ "step": 3230
2320
+ },
2321
+ {
2322
+ "epoch": 0.8244274809160306,
2323
+ "grad_norm": 0.6996018886566162,
2324
+ "learning_rate": 7.81453208934125e-06,
2325
+ "loss": 0.0025,
2326
+ "step": 3240
2327
+ },
2328
+ {
2329
+ "epoch": 0.8269720101781171,
2330
+ "grad_norm": 0.20250293612480164,
2331
+ "learning_rate": 7.701441899915184e-06,
2332
+ "loss": 0.0046,
2333
+ "step": 3250
2334
+ },
2335
+ {
2336
+ "epoch": 0.8295165394402035,
2337
+ "grad_norm": 1.6295372247695923,
2338
+ "learning_rate": 7.588351710489115e-06,
2339
+ "loss": 0.0018,
2340
+ "step": 3260
2341
+ },
2342
+ {
2343
+ "epoch": 0.8320610687022901,
2344
+ "grad_norm": 2.616621732711792,
2345
+ "learning_rate": 7.475261521063049e-06,
2346
+ "loss": 0.0074,
2347
+ "step": 3270
2348
+ },
2349
+ {
2350
+ "epoch": 0.8346055979643766,
2351
+ "grad_norm": 3.2906081676483154,
2352
+ "learning_rate": 7.362171331636981e-06,
2353
+ "loss": 0.0143,
2354
+ "step": 3280
2355
+ },
2356
+ {
2357
+ "epoch": 0.8371501272264631,
2358
+ "grad_norm": 1.3064147233963013,
2359
+ "learning_rate": 7.249081142210914e-06,
2360
+ "loss": 0.0056,
2361
+ "step": 3290
2362
+ },
2363
+ {
2364
+ "epoch": 0.8396946564885496,
2365
+ "grad_norm": 3.039973020553589,
2366
+ "learning_rate": 7.135990952784847e-06,
2367
+ "loss": 0.0095,
2368
+ "step": 3300
2369
+ },
2370
+ {
2371
+ "epoch": 0.8422391857506362,
2372
+ "grad_norm": 0.09989529848098755,
2373
+ "learning_rate": 7.022900763358779e-06,
2374
+ "loss": 0.0014,
2375
+ "step": 3310
2376
+ },
2377
+ {
2378
+ "epoch": 0.8447837150127226,
2379
+ "grad_norm": 0.2573828399181366,
2380
+ "learning_rate": 6.909810573932712e-06,
2381
+ "loss": 0.0061,
2382
+ "step": 3320
2383
+ },
2384
+ {
2385
+ "epoch": 0.8473282442748091,
2386
+ "grad_norm": 0.2958613932132721,
2387
+ "learning_rate": 6.796720384506644e-06,
2388
+ "loss": 0.004,
2389
+ "step": 3330
2390
+ },
2391
+ {
2392
+ "epoch": 0.8498727735368957,
2393
+ "grad_norm": 0.24500970542430878,
2394
+ "learning_rate": 6.683630195080577e-06,
2395
+ "loss": 0.0071,
2396
+ "step": 3340
2397
+ },
2398
+ {
2399
+ "epoch": 0.8524173027989822,
2400
+ "grad_norm": 1.2140284776687622,
2401
+ "learning_rate": 6.57054000565451e-06,
2402
+ "loss": 0.0036,
2403
+ "step": 3350
2404
+ },
2405
+ {
2406
+ "epoch": 0.8549618320610687,
2407
+ "grad_norm": 0.9599927067756653,
2408
+ "learning_rate": 6.457449816228443e-06,
2409
+ "loss": 0.0149,
2410
+ "step": 3360
2411
+ },
2412
+ {
2413
+ "epoch": 0.8575063613231552,
2414
+ "grad_norm": 0.27447447180747986,
2415
+ "learning_rate": 6.344359626802375e-06,
2416
+ "loss": 0.0074,
2417
+ "step": 3370
2418
+ },
2419
+ {
2420
+ "epoch": 0.8600508905852418,
2421
+ "grad_norm": 0.7029807567596436,
2422
+ "learning_rate": 6.231269437376308e-06,
2423
+ "loss": 0.007,
2424
+ "step": 3380
2425
+ },
2426
+ {
2427
+ "epoch": 0.8625954198473282,
2428
+ "grad_norm": 0.054117076098918915,
2429
+ "learning_rate": 6.118179247950241e-06,
2430
+ "loss": 0.0059,
2431
+ "step": 3390
2432
+ },
2433
+ {
2434
+ "epoch": 0.8651399491094147,
2435
+ "grad_norm": 0.4443982243537903,
2436
+ "learning_rate": 6.005089058524174e-06,
2437
+ "loss": 0.006,
2438
+ "step": 3400
2439
+ },
2440
+ {
2441
+ "epoch": 0.8676844783715013,
2442
+ "grad_norm": 3.201941967010498,
2443
+ "learning_rate": 5.8919988690981064e-06,
2444
+ "loss": 0.0118,
2445
+ "step": 3410
2446
+ },
2447
+ {
2448
+ "epoch": 0.8702290076335878,
2449
+ "grad_norm": 0.044088855385780334,
2450
+ "learning_rate": 5.778908679672038e-06,
2451
+ "loss": 0.0032,
2452
+ "step": 3420
2453
+ },
2454
+ {
2455
+ "epoch": 0.8727735368956743,
2456
+ "grad_norm": 1.8481858968734741,
2457
+ "learning_rate": 5.6658184902459716e-06,
2458
+ "loss": 0.0029,
2459
+ "step": 3430
2460
+ },
2461
+ {
2462
+ "epoch": 0.8753180661577609,
2463
+ "grad_norm": 0.014196817763149738,
2464
+ "learning_rate": 5.552728300819904e-06,
2465
+ "loss": 0.0062,
2466
+ "step": 3440
2467
+ },
2468
+ {
2469
+ "epoch": 0.8778625954198473,
2470
+ "grad_norm": 0.021374225616455078,
2471
+ "learning_rate": 5.439638111393837e-06,
2472
+ "loss": 0.0087,
2473
+ "step": 3450
2474
+ },
2475
+ {
2476
+ "epoch": 0.8804071246819338,
2477
+ "grad_norm": 0.0040234895423054695,
2478
+ "learning_rate": 5.326547921967769e-06,
2479
+ "loss": 0.0059,
2480
+ "step": 3460
2481
+ },
2482
+ {
2483
+ "epoch": 0.8829516539440203,
2484
+ "grad_norm": 0.07329881191253662,
2485
+ "learning_rate": 5.213457732541703e-06,
2486
+ "loss": 0.0023,
2487
+ "step": 3470
2488
+ },
2489
+ {
2490
+ "epoch": 0.8854961832061069,
2491
+ "grad_norm": 1.4617124795913696,
2492
+ "learning_rate": 5.100367543115635e-06,
2493
+ "loss": 0.0055,
2494
+ "step": 3480
2495
+ },
2496
+ {
2497
+ "epoch": 0.8880407124681934,
2498
+ "grad_norm": 2.9887943267822266,
2499
+ "learning_rate": 4.987277353689568e-06,
2500
+ "loss": 0.0106,
2501
+ "step": 3490
2502
+ },
2503
+ {
2504
+ "epoch": 0.8905852417302799,
2505
+ "grad_norm": 0.0617634542286396,
2506
+ "learning_rate": 4.8741871642635e-06,
2507
+ "loss": 0.0216,
2508
+ "step": 3500
2509
+ },
2510
+ {
2511
+ "epoch": 0.8905852417302799,
2512
+ "eval_loss": 0.02892346866428852,
2513
+ "eval_runtime": 101.0102,
2514
+ "eval_samples_per_second": 79.022,
2515
+ "eval_steps_per_second": 0.624,
2516
+ "step": 3500
2517
+ }
2518
+ ],
2519
+ "logging_steps": 10,
2520
+ "max_steps": 3930,
2521
+ "num_input_tokens_seen": 0,
2522
+ "num_train_epochs": 1,
2523
+ "save_steps": 500,
2524
+ "stateful_callbacks": {
2525
+ "TrainerControl": {
2526
+ "args": {
2527
+ "should_epoch_stop": false,
2528
+ "should_evaluate": false,
2529
+ "should_log": false,
2530
+ "should_save": true,
2531
+ "should_training_stop": false
2532
+ },
2533
+ "attributes": {}
2534
+ }
2535
+ },
2536
+ "total_flos": 1.9923556583082885e+18,
2537
+ "train_batch_size": 128,
2538
+ "trial_name": null,
2539
+ "trial_params": null
2540
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b80ed57aad323882c7df4384e7b8b2bc1f34dc2cf509d18264573453e1b2ebca
3
+ size 5905