bushali commited on
Commit
4f1554b
1 Parent(s): 11d992e

Upload 8 files

Browse files
README.md CHANGED
@@ -1,3 +1,202 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: HuggingFaceM4/idefics-9b
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 0.11.2.dev0
adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": {
4
+ "base_model_class": "IdeficsForVisionText2Text",
5
+ "parent_library": "transformers.models.idefics.modeling_idefics"
6
+ },
7
+ "base_model_name_or_path": "HuggingFaceM4/idefics-9b",
8
+ "bias": "none",
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 32,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 16,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "v_proj"
29
+ ],
30
+ "task_type": null,
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4adcdc1b27b26cc3b459237576a7c22af1a443ff6bf899f35702607e6ac96293
3
+ size 79071880
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77914a0be88e2bc63fb0df31962747962ac11f2fb3851b9ba4107e90e4d008dd
3
+ size 40107172
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5ada7e807b31cd7e67d730970b5eb49bbbb651c6fb8e146625ae362468f01e0
3
+ size 14244
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e4ab83e042e3deace3597b3caea23cdb552ba15ddeffd03637d82e011425dbf
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,2556 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9997069597069597,
5
+ "eval_steps": 100,
6
+ "global_step": 1706,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005860805860805861,
13
+ "grad_norm": 2.521568536758423,
14
+ "learning_rate": 0.00019941383352872216,
15
+ "loss": 2.3132,
16
+ "step": 5
17
+ },
18
+ {
19
+ "epoch": 0.011721611721611722,
20
+ "grad_norm": 2.104935884475708,
21
+ "learning_rate": 0.00019882766705744433,
22
+ "loss": 0.9444,
23
+ "step": 10
24
+ },
25
+ {
26
+ "epoch": 0.017582417582417582,
27
+ "grad_norm": 1.2950881719589233,
28
+ "learning_rate": 0.00019824150058616647,
29
+ "loss": 0.357,
30
+ "step": 15
31
+ },
32
+ {
33
+ "epoch": 0.023443223443223443,
34
+ "grad_norm": 1.1862170696258545,
35
+ "learning_rate": 0.00019765533411488865,
36
+ "loss": 0.2105,
37
+ "step": 20
38
+ },
39
+ {
40
+ "epoch": 0.029304029304029304,
41
+ "grad_norm": 0.5247148871421814,
42
+ "learning_rate": 0.0001970691676436108,
43
+ "loss": 0.1087,
44
+ "step": 25
45
+ },
46
+ {
47
+ "epoch": 0.035164835164835165,
48
+ "grad_norm": 1.2535285949707031,
49
+ "learning_rate": 0.00019648300117233296,
50
+ "loss": 0.1185,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.041025641025641026,
55
+ "grad_norm": 0.5407606959342957,
56
+ "learning_rate": 0.0001958968347010551,
57
+ "loss": 0.0775,
58
+ "step": 35
59
+ },
60
+ {
61
+ "epoch": 0.046886446886446886,
62
+ "grad_norm": 1.137266993522644,
63
+ "learning_rate": 0.00019531066822977726,
64
+ "loss": 0.0728,
65
+ "step": 40
66
+ },
67
+ {
68
+ "epoch": 0.05274725274725275,
69
+ "grad_norm": 0.5244052410125732,
70
+ "learning_rate": 0.00019472450175849943,
71
+ "loss": 0.056,
72
+ "step": 45
73
+ },
74
+ {
75
+ "epoch": 0.05860805860805861,
76
+ "grad_norm": 0.4511496126651764,
77
+ "learning_rate": 0.00019413833528722157,
78
+ "loss": 0.0648,
79
+ "step": 50
80
+ },
81
+ {
82
+ "epoch": 0.06446886446886448,
83
+ "grad_norm": 0.33913975954055786,
84
+ "learning_rate": 0.00019355216881594375,
85
+ "loss": 0.0479,
86
+ "step": 55
87
+ },
88
+ {
89
+ "epoch": 0.07032967032967033,
90
+ "grad_norm": 0.354777991771698,
91
+ "learning_rate": 0.0001929660023446659,
92
+ "loss": 0.0758,
93
+ "step": 60
94
+ },
95
+ {
96
+ "epoch": 0.0761904761904762,
97
+ "grad_norm": 0.3968910276889801,
98
+ "learning_rate": 0.00019237983587338807,
99
+ "loss": 0.0776,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 0.08205128205128205,
104
+ "grad_norm": 0.5061652660369873,
105
+ "learning_rate": 0.0001917936694021102,
106
+ "loss": 0.0349,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 0.08791208791208792,
111
+ "grad_norm": 0.24455250799655914,
112
+ "learning_rate": 0.00019120750293083236,
113
+ "loss": 0.0369,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 0.09377289377289377,
118
+ "grad_norm": 0.3959537446498871,
119
+ "learning_rate": 0.00019062133645955453,
120
+ "loss": 0.0513,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 0.09963369963369964,
125
+ "grad_norm": 0.4534469544887543,
126
+ "learning_rate": 0.00019003516998827668,
127
+ "loss": 0.0459,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 0.1054945054945055,
132
+ "grad_norm": 0.30694451928138733,
133
+ "learning_rate": 0.00018944900351699885,
134
+ "loss": 0.0377,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 0.11135531135531136,
139
+ "grad_norm": 0.15078052878379822,
140
+ "learning_rate": 0.000188862837045721,
141
+ "loss": 0.0297,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 0.11721611721611722,
146
+ "grad_norm": 0.3153330981731415,
147
+ "learning_rate": 0.00018827667057444317,
148
+ "loss": 0.0301,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 0.11721611721611722,
153
+ "eval_loss": 0.028654273599386215,
154
+ "eval_runtime": 4.8016,
155
+ "eval_samples_per_second": 5.831,
156
+ "eval_steps_per_second": 2.916,
157
+ "step": 100
158
+ },
159
+ {
160
+ "epoch": 0.12307692307692308,
161
+ "grad_norm": 0.5777396559715271,
162
+ "learning_rate": 0.0001876905041031653,
163
+ "loss": 0.0515,
164
+ "step": 105
165
+ },
166
+ {
167
+ "epoch": 0.12893772893772895,
168
+ "grad_norm": 0.3677718937397003,
169
+ "learning_rate": 0.00018710433763188746,
170
+ "loss": 0.0315,
171
+ "step": 110
172
+ },
173
+ {
174
+ "epoch": 0.1347985347985348,
175
+ "grad_norm": 0.30658507347106934,
176
+ "learning_rate": 0.0001865181711606096,
177
+ "loss": 0.0248,
178
+ "step": 115
179
+ },
180
+ {
181
+ "epoch": 0.14065934065934066,
182
+ "grad_norm": 0.21649648249149323,
183
+ "learning_rate": 0.00018593200468933178,
184
+ "loss": 0.0352,
185
+ "step": 120
186
+ },
187
+ {
188
+ "epoch": 0.14652014652014653,
189
+ "grad_norm": 0.2877885103225708,
190
+ "learning_rate": 0.00018534583821805395,
191
+ "loss": 0.0456,
192
+ "step": 125
193
+ },
194
+ {
195
+ "epoch": 0.1523809523809524,
196
+ "grad_norm": 0.1782904863357544,
197
+ "learning_rate": 0.0001847596717467761,
198
+ "loss": 0.0257,
199
+ "step": 130
200
+ },
201
+ {
202
+ "epoch": 0.15824175824175823,
203
+ "grad_norm": 0.1211300641298294,
204
+ "learning_rate": 0.00018417350527549827,
205
+ "loss": 0.0437,
206
+ "step": 135
207
+ },
208
+ {
209
+ "epoch": 0.1641025641025641,
210
+ "grad_norm": 0.298053503036499,
211
+ "learning_rate": 0.0001835873388042204,
212
+ "loss": 0.036,
213
+ "step": 140
214
+ },
215
+ {
216
+ "epoch": 0.16996336996336997,
217
+ "grad_norm": 0.2280658483505249,
218
+ "learning_rate": 0.00018300117233294256,
219
+ "loss": 0.0331,
220
+ "step": 145
221
+ },
222
+ {
223
+ "epoch": 0.17582417582417584,
224
+ "grad_norm": 0.14896267652511597,
225
+ "learning_rate": 0.0001824150058616647,
226
+ "loss": 0.0239,
227
+ "step": 150
228
+ },
229
+ {
230
+ "epoch": 0.18168498168498168,
231
+ "grad_norm": 0.1769961267709732,
232
+ "learning_rate": 0.00018182883939038688,
233
+ "loss": 0.0375,
234
+ "step": 155
235
+ },
236
+ {
237
+ "epoch": 0.18754578754578755,
238
+ "grad_norm": 0.19849297404289246,
239
+ "learning_rate": 0.00018124267291910902,
240
+ "loss": 0.0357,
241
+ "step": 160
242
+ },
243
+ {
244
+ "epoch": 0.1934065934065934,
245
+ "grad_norm": 0.2294420450925827,
246
+ "learning_rate": 0.0001806565064478312,
247
+ "loss": 0.0504,
248
+ "step": 165
249
+ },
250
+ {
251
+ "epoch": 0.19926739926739928,
252
+ "grad_norm": 0.10744224488735199,
253
+ "learning_rate": 0.00018007033997655337,
254
+ "loss": 0.0209,
255
+ "step": 170
256
+ },
257
+ {
258
+ "epoch": 0.20512820512820512,
259
+ "grad_norm": 0.06066066771745682,
260
+ "learning_rate": 0.00017948417350527551,
261
+ "loss": 0.0175,
262
+ "step": 175
263
+ },
264
+ {
265
+ "epoch": 0.210989010989011,
266
+ "grad_norm": 0.5421485304832458,
267
+ "learning_rate": 0.00017889800703399766,
268
+ "loss": 0.0398,
269
+ "step": 180
270
+ },
271
+ {
272
+ "epoch": 0.21684981684981686,
273
+ "grad_norm": 0.1306767612695694,
274
+ "learning_rate": 0.0001783118405627198,
275
+ "loss": 0.0258,
276
+ "step": 185
277
+ },
278
+ {
279
+ "epoch": 0.22271062271062272,
280
+ "grad_norm": 0.11510124802589417,
281
+ "learning_rate": 0.00017772567409144198,
282
+ "loss": 0.0253,
283
+ "step": 190
284
+ },
285
+ {
286
+ "epoch": 0.22857142857142856,
287
+ "grad_norm": 0.20577751100063324,
288
+ "learning_rate": 0.00017713950762016412,
289
+ "loss": 0.0277,
290
+ "step": 195
291
+ },
292
+ {
293
+ "epoch": 0.23443223443223443,
294
+ "grad_norm": 0.20588932931423187,
295
+ "learning_rate": 0.0001765533411488863,
296
+ "loss": 0.0296,
297
+ "step": 200
298
+ },
299
+ {
300
+ "epoch": 0.23443223443223443,
301
+ "eval_loss": 0.019306689500808716,
302
+ "eval_runtime": 4.7344,
303
+ "eval_samples_per_second": 5.914,
304
+ "eval_steps_per_second": 2.957,
305
+ "step": 200
306
+ },
307
+ {
308
+ "epoch": 0.2402930402930403,
309
+ "grad_norm": 0.13713975250720978,
310
+ "learning_rate": 0.00017596717467760847,
311
+ "loss": 0.0372,
312
+ "step": 205
313
+ },
314
+ {
315
+ "epoch": 0.24615384615384617,
316
+ "grad_norm": 0.14788508415222168,
317
+ "learning_rate": 0.00017538100820633061,
318
+ "loss": 0.033,
319
+ "step": 210
320
+ },
321
+ {
322
+ "epoch": 0.252014652014652,
323
+ "grad_norm": 0.2372630536556244,
324
+ "learning_rate": 0.00017479484173505276,
325
+ "loss": 0.0281,
326
+ "step": 215
327
+ },
328
+ {
329
+ "epoch": 0.2578754578754579,
330
+ "grad_norm": 0.2357954978942871,
331
+ "learning_rate": 0.0001742086752637749,
332
+ "loss": 0.0295,
333
+ "step": 220
334
+ },
335
+ {
336
+ "epoch": 0.26373626373626374,
337
+ "grad_norm": 0.2537606358528137,
338
+ "learning_rate": 0.00017362250879249708,
339
+ "loss": 0.036,
340
+ "step": 225
341
+ },
342
+ {
343
+ "epoch": 0.2695970695970696,
344
+ "grad_norm": 0.2222289741039276,
345
+ "learning_rate": 0.00017303634232121922,
346
+ "loss": 0.0402,
347
+ "step": 230
348
+ },
349
+ {
350
+ "epoch": 0.2754578754578755,
351
+ "grad_norm": 0.19696177542209625,
352
+ "learning_rate": 0.0001724501758499414,
353
+ "loss": 0.025,
354
+ "step": 235
355
+ },
356
+ {
357
+ "epoch": 0.2813186813186813,
358
+ "grad_norm": 0.08915109932422638,
359
+ "learning_rate": 0.00017186400937866357,
360
+ "loss": 0.0292,
361
+ "step": 240
362
+ },
363
+ {
364
+ "epoch": 0.28717948717948716,
365
+ "grad_norm": 0.39625948667526245,
366
+ "learning_rate": 0.00017127784290738572,
367
+ "loss": 0.0324,
368
+ "step": 245
369
+ },
370
+ {
371
+ "epoch": 0.29304029304029305,
372
+ "grad_norm": 0.05654177442193031,
373
+ "learning_rate": 0.00017069167643610786,
374
+ "loss": 0.0384,
375
+ "step": 250
376
+ },
377
+ {
378
+ "epoch": 0.2989010989010989,
379
+ "grad_norm": 0.23707075417041779,
380
+ "learning_rate": 0.00017010550996483,
381
+ "loss": 0.0311,
382
+ "step": 255
383
+ },
384
+ {
385
+ "epoch": 0.3047619047619048,
386
+ "grad_norm": 0.2619571387767792,
387
+ "learning_rate": 0.00016951934349355218,
388
+ "loss": 0.0275,
389
+ "step": 260
390
+ },
391
+ {
392
+ "epoch": 0.31062271062271063,
393
+ "grad_norm": 0.11028550565242767,
394
+ "learning_rate": 0.00016893317702227432,
395
+ "loss": 0.0194,
396
+ "step": 265
397
+ },
398
+ {
399
+ "epoch": 0.31648351648351647,
400
+ "grad_norm": 0.26710912585258484,
401
+ "learning_rate": 0.0001683470105509965,
402
+ "loss": 0.0318,
403
+ "step": 270
404
+ },
405
+ {
406
+ "epoch": 0.32234432234432236,
407
+ "grad_norm": 0.20064710080623627,
408
+ "learning_rate": 0.00016776084407971864,
409
+ "loss": 0.0517,
410
+ "step": 275
411
+ },
412
+ {
413
+ "epoch": 0.3282051282051282,
414
+ "grad_norm": 0.06760745495557785,
415
+ "learning_rate": 0.00016717467760844082,
416
+ "loss": 0.0223,
417
+ "step": 280
418
+ },
419
+ {
420
+ "epoch": 0.33406593406593404,
421
+ "grad_norm": 0.14518442749977112,
422
+ "learning_rate": 0.00016658851113716296,
423
+ "loss": 0.0216,
424
+ "step": 285
425
+ },
426
+ {
427
+ "epoch": 0.33992673992673994,
428
+ "grad_norm": 0.35427016019821167,
429
+ "learning_rate": 0.0001660023446658851,
430
+ "loss": 0.0268,
431
+ "step": 290
432
+ },
433
+ {
434
+ "epoch": 0.3457875457875458,
435
+ "grad_norm": 0.14693213999271393,
436
+ "learning_rate": 0.00016541617819460728,
437
+ "loss": 0.0246,
438
+ "step": 295
439
+ },
440
+ {
441
+ "epoch": 0.3516483516483517,
442
+ "grad_norm": 0.2351713478565216,
443
+ "learning_rate": 0.00016483001172332943,
444
+ "loss": 0.0399,
445
+ "step": 300
446
+ },
447
+ {
448
+ "epoch": 0.3516483516483517,
449
+ "eval_loss": 0.02264154888689518,
450
+ "eval_runtime": 4.8013,
451
+ "eval_samples_per_second": 5.832,
452
+ "eval_steps_per_second": 2.916,
453
+ "step": 300
454
+ },
455
+ {
456
+ "epoch": 0.3575091575091575,
457
+ "grad_norm": 0.16985514760017395,
458
+ "learning_rate": 0.0001642438452520516,
459
+ "loss": 0.0243,
460
+ "step": 305
461
+ },
462
+ {
463
+ "epoch": 0.36336996336996336,
464
+ "grad_norm": 0.199642613530159,
465
+ "learning_rate": 0.00016365767878077374,
466
+ "loss": 0.0329,
467
+ "step": 310
468
+ },
469
+ {
470
+ "epoch": 0.36923076923076925,
471
+ "grad_norm": 0.05538804084062576,
472
+ "learning_rate": 0.00016307151230949592,
473
+ "loss": 0.0474,
474
+ "step": 315
475
+ },
476
+ {
477
+ "epoch": 0.3750915750915751,
478
+ "grad_norm": 0.1542443037033081,
479
+ "learning_rate": 0.00016248534583821806,
480
+ "loss": 0.0232,
481
+ "step": 320
482
+ },
483
+ {
484
+ "epoch": 0.38095238095238093,
485
+ "grad_norm": 0.16808335483074188,
486
+ "learning_rate": 0.0001618991793669402,
487
+ "loss": 0.0272,
488
+ "step": 325
489
+ },
490
+ {
491
+ "epoch": 0.3868131868131868,
492
+ "grad_norm": 0.20816853642463684,
493
+ "learning_rate": 0.00016131301289566238,
494
+ "loss": 0.0375,
495
+ "step": 330
496
+ },
497
+ {
498
+ "epoch": 0.39267399267399267,
499
+ "grad_norm": 0.10338038206100464,
500
+ "learning_rate": 0.00016072684642438453,
501
+ "loss": 0.0233,
502
+ "step": 335
503
+ },
504
+ {
505
+ "epoch": 0.39853479853479856,
506
+ "grad_norm": 0.06522126495838165,
507
+ "learning_rate": 0.0001601406799531067,
508
+ "loss": 0.0288,
509
+ "step": 340
510
+ },
511
+ {
512
+ "epoch": 0.4043956043956044,
513
+ "grad_norm": 0.11830403655767441,
514
+ "learning_rate": 0.00015955451348182884,
515
+ "loss": 0.0164,
516
+ "step": 345
517
+ },
518
+ {
519
+ "epoch": 0.41025641025641024,
520
+ "grad_norm": 0.2777006924152374,
521
+ "learning_rate": 0.00015896834701055102,
522
+ "loss": 0.0345,
523
+ "step": 350
524
+ },
525
+ {
526
+ "epoch": 0.41611721611721614,
527
+ "grad_norm": 0.1277918815612793,
528
+ "learning_rate": 0.00015838218053927316,
529
+ "loss": 0.0229,
530
+ "step": 355
531
+ },
532
+ {
533
+ "epoch": 0.421978021978022,
534
+ "grad_norm": 0.09861145913600922,
535
+ "learning_rate": 0.0001577960140679953,
536
+ "loss": 0.0181,
537
+ "step": 360
538
+ },
539
+ {
540
+ "epoch": 0.4278388278388278,
541
+ "grad_norm": 0.08698171377182007,
542
+ "learning_rate": 0.00015720984759671748,
543
+ "loss": 0.0365,
544
+ "step": 365
545
+ },
546
+ {
547
+ "epoch": 0.4336996336996337,
548
+ "grad_norm": 0.23488883674144745,
549
+ "learning_rate": 0.00015662368112543963,
550
+ "loss": 0.0352,
551
+ "step": 370
552
+ },
553
+ {
554
+ "epoch": 0.43956043956043955,
555
+ "grad_norm": 0.05140375718474388,
556
+ "learning_rate": 0.0001560375146541618,
557
+ "loss": 0.0235,
558
+ "step": 375
559
+ },
560
+ {
561
+ "epoch": 0.44542124542124545,
562
+ "grad_norm": 0.15759135782718658,
563
+ "learning_rate": 0.00015545134818288395,
564
+ "loss": 0.0236,
565
+ "step": 380
566
+ },
567
+ {
568
+ "epoch": 0.4512820512820513,
569
+ "grad_norm": 0.07523085922002792,
570
+ "learning_rate": 0.00015486518171160612,
571
+ "loss": 0.0218,
572
+ "step": 385
573
+ },
574
+ {
575
+ "epoch": 0.45714285714285713,
576
+ "grad_norm": 0.1892630159854889,
577
+ "learning_rate": 0.00015427901524032826,
578
+ "loss": 0.0264,
579
+ "step": 390
580
+ },
581
+ {
582
+ "epoch": 0.463003663003663,
583
+ "grad_norm": 0.12749487161636353,
584
+ "learning_rate": 0.0001536928487690504,
585
+ "loss": 0.0414,
586
+ "step": 395
587
+ },
588
+ {
589
+ "epoch": 0.46886446886446886,
590
+ "grad_norm": 0.11519593745470047,
591
+ "learning_rate": 0.00015310668229777258,
592
+ "loss": 0.0173,
593
+ "step": 400
594
+ },
595
+ {
596
+ "epoch": 0.46886446886446886,
597
+ "eval_loss": 0.01765686459839344,
598
+ "eval_runtime": 4.8182,
599
+ "eval_samples_per_second": 5.811,
600
+ "eval_steps_per_second": 2.906,
601
+ "step": 400
602
+ },
603
+ {
604
+ "epoch": 0.4747252747252747,
605
+ "grad_norm": 0.14240577816963196,
606
+ "learning_rate": 0.00015252051582649473,
607
+ "loss": 0.0214,
608
+ "step": 405
609
+ },
610
+ {
611
+ "epoch": 0.4805860805860806,
612
+ "grad_norm": 0.11283282935619354,
613
+ "learning_rate": 0.0001519343493552169,
614
+ "loss": 0.0266,
615
+ "step": 410
616
+ },
617
+ {
618
+ "epoch": 0.48644688644688644,
619
+ "grad_norm": 0.16279707849025726,
620
+ "learning_rate": 0.00015134818288393905,
621
+ "loss": 0.0274,
622
+ "step": 415
623
+ },
624
+ {
625
+ "epoch": 0.49230769230769234,
626
+ "grad_norm": 0.07319923490285873,
627
+ "learning_rate": 0.00015076201641266122,
628
+ "loss": 0.0216,
629
+ "step": 420
630
+ },
631
+ {
632
+ "epoch": 0.4981684981684982,
633
+ "grad_norm": 0.13532328605651855,
634
+ "learning_rate": 0.00015017584994138336,
635
+ "loss": 0.0308,
636
+ "step": 425
637
+ },
638
+ {
639
+ "epoch": 0.504029304029304,
640
+ "grad_norm": 0.12896127998828888,
641
+ "learning_rate": 0.0001495896834701055,
642
+ "loss": 0.0129,
643
+ "step": 430
644
+ },
645
+ {
646
+ "epoch": 0.5098901098901099,
647
+ "grad_norm": 0.03718201071023941,
648
+ "learning_rate": 0.00014900351699882766,
649
+ "loss": 0.0181,
650
+ "step": 435
651
+ },
652
+ {
653
+ "epoch": 0.5157509157509158,
654
+ "grad_norm": 0.05926808714866638,
655
+ "learning_rate": 0.00014841735052754983,
656
+ "loss": 0.0146,
657
+ "step": 440
658
+ },
659
+ {
660
+ "epoch": 0.5216117216117216,
661
+ "grad_norm": 0.09976931661367416,
662
+ "learning_rate": 0.00014783118405627197,
663
+ "loss": 0.0231,
664
+ "step": 445
665
+ },
666
+ {
667
+ "epoch": 0.5274725274725275,
668
+ "grad_norm": 0.18460391461849213,
669
+ "learning_rate": 0.00014724501758499415,
670
+ "loss": 0.02,
671
+ "step": 450
672
+ },
673
+ {
674
+ "epoch": 0.5333333333333333,
675
+ "grad_norm": 0.07457377016544342,
676
+ "learning_rate": 0.00014665885111371632,
677
+ "loss": 0.0209,
678
+ "step": 455
679
+ },
680
+ {
681
+ "epoch": 0.5391941391941392,
682
+ "grad_norm": 0.03992030769586563,
683
+ "learning_rate": 0.00014607268464243847,
684
+ "loss": 0.0251,
685
+ "step": 460
686
+ },
687
+ {
688
+ "epoch": 0.545054945054945,
689
+ "grad_norm": 0.196414053440094,
690
+ "learning_rate": 0.0001454865181711606,
691
+ "loss": 0.0344,
692
+ "step": 465
693
+ },
694
+ {
695
+ "epoch": 0.550915750915751,
696
+ "grad_norm": 0.19978556036949158,
697
+ "learning_rate": 0.00014490035169988276,
698
+ "loss": 0.0159,
699
+ "step": 470
700
+ },
701
+ {
702
+ "epoch": 0.5567765567765568,
703
+ "grad_norm": 0.11389517784118652,
704
+ "learning_rate": 0.00014431418522860493,
705
+ "loss": 0.0197,
706
+ "step": 475
707
+ },
708
+ {
709
+ "epoch": 0.5626373626373626,
710
+ "grad_norm": 0.07047716528177261,
711
+ "learning_rate": 0.00014372801875732708,
712
+ "loss": 0.0138,
713
+ "step": 480
714
+ },
715
+ {
716
+ "epoch": 0.5684981684981685,
717
+ "grad_norm": 0.10614708811044693,
718
+ "learning_rate": 0.00014314185228604925,
719
+ "loss": 0.0208,
720
+ "step": 485
721
+ },
722
+ {
723
+ "epoch": 0.5743589743589743,
724
+ "grad_norm": 0.20569799840450287,
725
+ "learning_rate": 0.00014255568581477142,
726
+ "loss": 0.0203,
727
+ "step": 490
728
+ },
729
+ {
730
+ "epoch": 0.5802197802197803,
731
+ "grad_norm": 0.19309553503990173,
732
+ "learning_rate": 0.00014196951934349357,
733
+ "loss": 0.0282,
734
+ "step": 495
735
+ },
736
+ {
737
+ "epoch": 0.5860805860805861,
738
+ "grad_norm": 0.07542768865823746,
739
+ "learning_rate": 0.0001413833528722157,
740
+ "loss": 0.0173,
741
+ "step": 500
742
+ },
743
+ {
744
+ "epoch": 0.5860805860805861,
745
+ "eval_loss": 0.022126102820038795,
746
+ "eval_runtime": 4.8755,
747
+ "eval_samples_per_second": 5.743,
748
+ "eval_steps_per_second": 2.872,
749
+ "step": 500
750
+ },
751
+ {
752
+ "epoch": 0.591941391941392,
753
+ "grad_norm": 0.1205630674958229,
754
+ "learning_rate": 0.00014079718640093786,
755
+ "loss": 0.0322,
756
+ "step": 505
757
+ },
758
+ {
759
+ "epoch": 0.5978021978021978,
760
+ "grad_norm": 0.13761042058467865,
761
+ "learning_rate": 0.00014021101992966003,
762
+ "loss": 0.0203,
763
+ "step": 510
764
+ },
765
+ {
766
+ "epoch": 0.6036630036630036,
767
+ "grad_norm": 0.08595598489046097,
768
+ "learning_rate": 0.00013962485345838218,
769
+ "loss": 0.0145,
770
+ "step": 515
771
+ },
772
+ {
773
+ "epoch": 0.6095238095238096,
774
+ "grad_norm": 0.11087319999933243,
775
+ "learning_rate": 0.00013903868698710435,
776
+ "loss": 0.0218,
777
+ "step": 520
778
+ },
779
+ {
780
+ "epoch": 0.6153846153846154,
781
+ "grad_norm": 0.14962054789066315,
782
+ "learning_rate": 0.00013845252051582652,
783
+ "loss": 0.0322,
784
+ "step": 525
785
+ },
786
+ {
787
+ "epoch": 0.6212454212454213,
788
+ "grad_norm": 0.07573894411325455,
789
+ "learning_rate": 0.00013786635404454867,
790
+ "loss": 0.0275,
791
+ "step": 530
792
+ },
793
+ {
794
+ "epoch": 0.6271062271062271,
795
+ "grad_norm": 0.069780133664608,
796
+ "learning_rate": 0.0001372801875732708,
797
+ "loss": 0.0235,
798
+ "step": 535
799
+ },
800
+ {
801
+ "epoch": 0.6329670329670329,
802
+ "grad_norm": 0.07833613455295563,
803
+ "learning_rate": 0.00013669402110199296,
804
+ "loss": 0.0344,
805
+ "step": 540
806
+ },
807
+ {
808
+ "epoch": 0.6388278388278388,
809
+ "grad_norm": 0.07331829518079758,
810
+ "learning_rate": 0.00013610785463071513,
811
+ "loss": 0.0135,
812
+ "step": 545
813
+ },
814
+ {
815
+ "epoch": 0.6446886446886447,
816
+ "grad_norm": 0.22369089722633362,
817
+ "learning_rate": 0.00013552168815943728,
818
+ "loss": 0.0222,
819
+ "step": 550
820
+ },
821
+ {
822
+ "epoch": 0.6505494505494506,
823
+ "grad_norm": 0.1465146392583847,
824
+ "learning_rate": 0.00013493552168815945,
825
+ "loss": 0.0307,
826
+ "step": 555
827
+ },
828
+ {
829
+ "epoch": 0.6564102564102564,
830
+ "grad_norm": 0.06348715722560883,
831
+ "learning_rate": 0.00013434935521688162,
832
+ "loss": 0.0358,
833
+ "step": 560
834
+ },
835
+ {
836
+ "epoch": 0.6622710622710622,
837
+ "grad_norm": 0.09298256784677505,
838
+ "learning_rate": 0.00013376318874560377,
839
+ "loss": 0.0224,
840
+ "step": 565
841
+ },
842
+ {
843
+ "epoch": 0.6681318681318681,
844
+ "grad_norm": 0.18280836939811707,
845
+ "learning_rate": 0.00013317702227432591,
846
+ "loss": 0.0263,
847
+ "step": 570
848
+ },
849
+ {
850
+ "epoch": 0.673992673992674,
851
+ "grad_norm": 0.07080171257257462,
852
+ "learning_rate": 0.00013259085580304806,
853
+ "loss": 0.0192,
854
+ "step": 575
855
+ },
856
+ {
857
+ "epoch": 0.6798534798534799,
858
+ "grad_norm": 0.11019092798233032,
859
+ "learning_rate": 0.00013200468933177023,
860
+ "loss": 0.0211,
861
+ "step": 580
862
+ },
863
+ {
864
+ "epoch": 0.6857142857142857,
865
+ "grad_norm": 0.13162659108638763,
866
+ "learning_rate": 0.00013141852286049238,
867
+ "loss": 0.0284,
868
+ "step": 585
869
+ },
870
+ {
871
+ "epoch": 0.6915750915750916,
872
+ "grad_norm": 0.19205588102340698,
873
+ "learning_rate": 0.00013083235638921455,
874
+ "loss": 0.0245,
875
+ "step": 590
876
+ },
877
+ {
878
+ "epoch": 0.6974358974358974,
879
+ "grad_norm": 0.0971173569560051,
880
+ "learning_rate": 0.0001302461899179367,
881
+ "loss": 0.0216,
882
+ "step": 595
883
+ },
884
+ {
885
+ "epoch": 0.7032967032967034,
886
+ "grad_norm": 0.2179749310016632,
887
+ "learning_rate": 0.00012966002344665887,
888
+ "loss": 0.0268,
889
+ "step": 600
890
+ },
891
+ {
892
+ "epoch": 0.7032967032967034,
893
+ "eval_loss": 0.01850169710814953,
894
+ "eval_runtime": 4.6688,
895
+ "eval_samples_per_second": 5.997,
896
+ "eval_steps_per_second": 2.999,
897
+ "step": 600
898
+ },
899
+ {
900
+ "epoch": 0.7091575091575092,
901
+ "grad_norm": 0.23079490661621094,
902
+ "learning_rate": 0.00012907385697538101,
903
+ "loss": 0.0269,
904
+ "step": 605
905
+ },
906
+ {
907
+ "epoch": 0.715018315018315,
908
+ "grad_norm": 0.08325810730457306,
909
+ "learning_rate": 0.00012848769050410316,
910
+ "loss": 0.0211,
911
+ "step": 610
912
+ },
913
+ {
914
+ "epoch": 0.7208791208791209,
915
+ "grad_norm": 0.05983910337090492,
916
+ "learning_rate": 0.00012790152403282533,
917
+ "loss": 0.0217,
918
+ "step": 615
919
+ },
920
+ {
921
+ "epoch": 0.7267399267399267,
922
+ "grad_norm": 0.13831888139247894,
923
+ "learning_rate": 0.00012731535756154748,
924
+ "loss": 0.0142,
925
+ "step": 620
926
+ },
927
+ {
928
+ "epoch": 0.7326007326007326,
929
+ "grad_norm": 0.12296965718269348,
930
+ "learning_rate": 0.00012672919109026965,
931
+ "loss": 0.0253,
932
+ "step": 625
933
+ },
934
+ {
935
+ "epoch": 0.7384615384615385,
936
+ "grad_norm": 0.13777951896190643,
937
+ "learning_rate": 0.0001261430246189918,
938
+ "loss": 0.0214,
939
+ "step": 630
940
+ },
941
+ {
942
+ "epoch": 0.7443223443223443,
943
+ "grad_norm": 0.12136834859848022,
944
+ "learning_rate": 0.00012555685814771397,
945
+ "loss": 0.0244,
946
+ "step": 635
947
+ },
948
+ {
949
+ "epoch": 0.7501831501831502,
950
+ "grad_norm": 0.050576552748680115,
951
+ "learning_rate": 0.00012497069167643612,
952
+ "loss": 0.0137,
953
+ "step": 640
954
+ },
955
+ {
956
+ "epoch": 0.756043956043956,
957
+ "grad_norm": 0.22222141921520233,
958
+ "learning_rate": 0.00012438452520515826,
959
+ "loss": 0.0254,
960
+ "step": 645
961
+ },
962
+ {
963
+ "epoch": 0.7619047619047619,
964
+ "grad_norm": 0.06815624237060547,
965
+ "learning_rate": 0.00012379835873388043,
966
+ "loss": 0.0231,
967
+ "step": 650
968
+ },
969
+ {
970
+ "epoch": 0.7677655677655678,
971
+ "grad_norm": 0.19518345594406128,
972
+ "learning_rate": 0.00012321219226260258,
973
+ "loss": 0.0218,
974
+ "step": 655
975
+ },
976
+ {
977
+ "epoch": 0.7736263736263737,
978
+ "grad_norm": 0.06349798291921616,
979
+ "learning_rate": 0.00012262602579132475,
980
+ "loss": 0.0265,
981
+ "step": 660
982
+ },
983
+ {
984
+ "epoch": 0.7794871794871795,
985
+ "grad_norm": 0.09031341969966888,
986
+ "learning_rate": 0.00012203985932004688,
987
+ "loss": 0.0299,
988
+ "step": 665
989
+ },
990
+ {
991
+ "epoch": 0.7853479853479853,
992
+ "grad_norm": 0.060232892632484436,
993
+ "learning_rate": 0.00012145369284876906,
994
+ "loss": 0.0227,
995
+ "step": 670
996
+ },
997
+ {
998
+ "epoch": 0.7912087912087912,
999
+ "grad_norm": 0.23972396552562714,
1000
+ "learning_rate": 0.00012086752637749122,
1001
+ "loss": 0.0233,
1002
+ "step": 675
1003
+ },
1004
+ {
1005
+ "epoch": 0.7970695970695971,
1006
+ "grad_norm": 0.06141636520624161,
1007
+ "learning_rate": 0.00012028135990621336,
1008
+ "loss": 0.017,
1009
+ "step": 680
1010
+ },
1011
+ {
1012
+ "epoch": 0.802930402930403,
1013
+ "grad_norm": 0.05603253096342087,
1014
+ "learning_rate": 0.00011969519343493553,
1015
+ "loss": 0.025,
1016
+ "step": 685
1017
+ },
1018
+ {
1019
+ "epoch": 0.8087912087912088,
1020
+ "grad_norm": 0.06840907782316208,
1021
+ "learning_rate": 0.0001191090269636577,
1022
+ "loss": 0.0164,
1023
+ "step": 690
1024
+ },
1025
+ {
1026
+ "epoch": 0.8146520146520146,
1027
+ "grad_norm": 0.1270790845155716,
1028
+ "learning_rate": 0.00011852286049237984,
1029
+ "loss": 0.0237,
1030
+ "step": 695
1031
+ },
1032
+ {
1033
+ "epoch": 0.8205128205128205,
1034
+ "grad_norm": 0.03222518041729927,
1035
+ "learning_rate": 0.00011793669402110198,
1036
+ "loss": 0.0169,
1037
+ "step": 700
1038
+ },
1039
+ {
1040
+ "epoch": 0.8205128205128205,
1041
+ "eval_loss": 0.026108432561159134,
1042
+ "eval_runtime": 4.6865,
1043
+ "eval_samples_per_second": 5.975,
1044
+ "eval_steps_per_second": 2.987,
1045
+ "step": 700
1046
+ },
1047
+ {
1048
+ "epoch": 0.8263736263736263,
1049
+ "grad_norm": 0.06273896247148514,
1050
+ "learning_rate": 0.00011735052754982416,
1051
+ "loss": 0.0245,
1052
+ "step": 705
1053
+ },
1054
+ {
1055
+ "epoch": 0.8322344322344323,
1056
+ "grad_norm": 0.1422451138496399,
1057
+ "learning_rate": 0.00011676436107854632,
1058
+ "loss": 0.0218,
1059
+ "step": 710
1060
+ },
1061
+ {
1062
+ "epoch": 0.8380952380952381,
1063
+ "grad_norm": 0.07166247069835663,
1064
+ "learning_rate": 0.00011617819460726846,
1065
+ "loss": 0.0259,
1066
+ "step": 715
1067
+ },
1068
+ {
1069
+ "epoch": 0.843956043956044,
1070
+ "grad_norm": 0.13188450038433075,
1071
+ "learning_rate": 0.00011559202813599064,
1072
+ "loss": 0.0185,
1073
+ "step": 720
1074
+ },
1075
+ {
1076
+ "epoch": 0.8498168498168498,
1077
+ "grad_norm": 0.11839079111814499,
1078
+ "learning_rate": 0.0001150058616647128,
1079
+ "loss": 0.0196,
1080
+ "step": 725
1081
+ },
1082
+ {
1083
+ "epoch": 0.8556776556776556,
1084
+ "grad_norm": 0.09421879053115845,
1085
+ "learning_rate": 0.00011441969519343494,
1086
+ "loss": 0.0207,
1087
+ "step": 730
1088
+ },
1089
+ {
1090
+ "epoch": 0.8615384615384616,
1091
+ "grad_norm": 0.11722107976675034,
1092
+ "learning_rate": 0.00011383352872215709,
1093
+ "loss": 0.0286,
1094
+ "step": 735
1095
+ },
1096
+ {
1097
+ "epoch": 0.8673992673992674,
1098
+ "grad_norm": 0.07790110260248184,
1099
+ "learning_rate": 0.00011324736225087926,
1100
+ "loss": 0.0157,
1101
+ "step": 740
1102
+ },
1103
+ {
1104
+ "epoch": 0.8732600732600733,
1105
+ "grad_norm": 0.11153840273618698,
1106
+ "learning_rate": 0.00011266119577960142,
1107
+ "loss": 0.0184,
1108
+ "step": 745
1109
+ },
1110
+ {
1111
+ "epoch": 0.8791208791208791,
1112
+ "grad_norm": 0.07105362415313721,
1113
+ "learning_rate": 0.00011207502930832356,
1114
+ "loss": 0.0193,
1115
+ "step": 750
1116
+ },
1117
+ {
1118
+ "epoch": 0.884981684981685,
1119
+ "grad_norm": 0.11616308242082596,
1120
+ "learning_rate": 0.00011148886283704571,
1121
+ "loss": 0.0219,
1122
+ "step": 755
1123
+ },
1124
+ {
1125
+ "epoch": 0.8908424908424909,
1126
+ "grad_norm": 0.10045047104358673,
1127
+ "learning_rate": 0.0001109026963657679,
1128
+ "loss": 0.0177,
1129
+ "step": 760
1130
+ },
1131
+ {
1132
+ "epoch": 0.8967032967032967,
1133
+ "grad_norm": 0.07033990323543549,
1134
+ "learning_rate": 0.00011031652989449004,
1135
+ "loss": 0.0227,
1136
+ "step": 765
1137
+ },
1138
+ {
1139
+ "epoch": 0.9025641025641026,
1140
+ "grad_norm": 0.07648850232362747,
1141
+ "learning_rate": 0.00010973036342321219,
1142
+ "loss": 0.023,
1143
+ "step": 770
1144
+ },
1145
+ {
1146
+ "epoch": 0.9084249084249084,
1147
+ "grad_norm": 0.05392804369330406,
1148
+ "learning_rate": 0.00010914419695193436,
1149
+ "loss": 0.0136,
1150
+ "step": 775
1151
+ },
1152
+ {
1153
+ "epoch": 0.9142857142857143,
1154
+ "grad_norm": 0.17311276495456696,
1155
+ "learning_rate": 0.00010855803048065652,
1156
+ "loss": 0.0257,
1157
+ "step": 780
1158
+ },
1159
+ {
1160
+ "epoch": 0.9201465201465201,
1161
+ "grad_norm": 0.07022574543952942,
1162
+ "learning_rate": 0.00010797186400937866,
1163
+ "loss": 0.0282,
1164
+ "step": 785
1165
+ },
1166
+ {
1167
+ "epoch": 0.926007326007326,
1168
+ "grad_norm": 0.15858297049999237,
1169
+ "learning_rate": 0.00010738569753810081,
1170
+ "loss": 0.0219,
1171
+ "step": 790
1172
+ },
1173
+ {
1174
+ "epoch": 0.9318681318681319,
1175
+ "grad_norm": 0.06796769052743912,
1176
+ "learning_rate": 0.00010679953106682298,
1177
+ "loss": 0.0288,
1178
+ "step": 795
1179
+ },
1180
+ {
1181
+ "epoch": 0.9377289377289377,
1182
+ "grad_norm": 0.11868051439523697,
1183
+ "learning_rate": 0.00010621336459554514,
1184
+ "loss": 0.0248,
1185
+ "step": 800
1186
+ },
1187
+ {
1188
+ "epoch": 0.9377289377289377,
1189
+ "eval_loss": 0.019950957968831062,
1190
+ "eval_runtime": 4.8291,
1191
+ "eval_samples_per_second": 5.798,
1192
+ "eval_steps_per_second": 2.899,
1193
+ "step": 800
1194
+ },
1195
+ {
1196
+ "epoch": 0.9435897435897436,
1197
+ "grad_norm": 0.19787561893463135,
1198
+ "learning_rate": 0.00010562719812426729,
1199
+ "loss": 0.0249,
1200
+ "step": 805
1201
+ },
1202
+ {
1203
+ "epoch": 0.9494505494505494,
1204
+ "grad_norm": 0.06437662243843079,
1205
+ "learning_rate": 0.00010504103165298946,
1206
+ "loss": 0.0141,
1207
+ "step": 810
1208
+ },
1209
+ {
1210
+ "epoch": 0.9553113553113554,
1211
+ "grad_norm": 0.09178975969552994,
1212
+ "learning_rate": 0.00010445486518171162,
1213
+ "loss": 0.0218,
1214
+ "step": 815
1215
+ },
1216
+ {
1217
+ "epoch": 0.9611721611721612,
1218
+ "grad_norm": 0.09567834436893463,
1219
+ "learning_rate": 0.00010386869871043376,
1220
+ "loss": 0.0229,
1221
+ "step": 820
1222
+ },
1223
+ {
1224
+ "epoch": 0.967032967032967,
1225
+ "grad_norm": 0.039594829082489014,
1226
+ "learning_rate": 0.00010328253223915591,
1227
+ "loss": 0.0186,
1228
+ "step": 825
1229
+ },
1230
+ {
1231
+ "epoch": 0.9728937728937729,
1232
+ "grad_norm": 0.18495650589466095,
1233
+ "learning_rate": 0.00010269636576787808,
1234
+ "loss": 0.0237,
1235
+ "step": 830
1236
+ },
1237
+ {
1238
+ "epoch": 0.9787545787545787,
1239
+ "grad_norm": 0.1861388385295868,
1240
+ "learning_rate": 0.00010211019929660024,
1241
+ "loss": 0.0367,
1242
+ "step": 835
1243
+ },
1244
+ {
1245
+ "epoch": 0.9846153846153847,
1246
+ "grad_norm": 0.05491223558783531,
1247
+ "learning_rate": 0.00010152403282532239,
1248
+ "loss": 0.015,
1249
+ "step": 840
1250
+ },
1251
+ {
1252
+ "epoch": 0.9904761904761905,
1253
+ "grad_norm": 0.04110349714756012,
1254
+ "learning_rate": 0.00010093786635404456,
1255
+ "loss": 0.0258,
1256
+ "step": 845
1257
+ },
1258
+ {
1259
+ "epoch": 0.9963369963369964,
1260
+ "grad_norm": 0.07649147510528564,
1261
+ "learning_rate": 0.00010035169988276672,
1262
+ "loss": 0.0235,
1263
+ "step": 850
1264
+ },
1265
+ {
1266
+ "epoch": 1.0021978021978022,
1267
+ "grad_norm": 0.11078579723834991,
1268
+ "learning_rate": 9.976553341148887e-05,
1269
+ "loss": 0.0204,
1270
+ "step": 855
1271
+ },
1272
+ {
1273
+ "epoch": 1.008058608058608,
1274
+ "grad_norm": 0.08302613347768784,
1275
+ "learning_rate": 9.917936694021102e-05,
1276
+ "loss": 0.0188,
1277
+ "step": 860
1278
+ },
1279
+ {
1280
+ "epoch": 1.0139194139194139,
1281
+ "grad_norm": 0.19045108556747437,
1282
+ "learning_rate": 9.859320046893318e-05,
1283
+ "loss": 0.0226,
1284
+ "step": 865
1285
+ },
1286
+ {
1287
+ "epoch": 1.0197802197802197,
1288
+ "grad_norm": 0.04657626897096634,
1289
+ "learning_rate": 9.800703399765534e-05,
1290
+ "loss": 0.0205,
1291
+ "step": 870
1292
+ },
1293
+ {
1294
+ "epoch": 1.0256410256410255,
1295
+ "grad_norm": 0.10237232595682144,
1296
+ "learning_rate": 9.742086752637749e-05,
1297
+ "loss": 0.0259,
1298
+ "step": 875
1299
+ },
1300
+ {
1301
+ "epoch": 1.0315018315018314,
1302
+ "grad_norm": 0.1746947020292282,
1303
+ "learning_rate": 9.683470105509965e-05,
1304
+ "loss": 0.0229,
1305
+ "step": 880
1306
+ },
1307
+ {
1308
+ "epoch": 1.0373626373626375,
1309
+ "grad_norm": 0.1032433807849884,
1310
+ "learning_rate": 9.624853458382182e-05,
1311
+ "loss": 0.0277,
1312
+ "step": 885
1313
+ },
1314
+ {
1315
+ "epoch": 1.0432234432234433,
1316
+ "grad_norm": 0.08428288996219635,
1317
+ "learning_rate": 9.566236811254397e-05,
1318
+ "loss": 0.0161,
1319
+ "step": 890
1320
+ },
1321
+ {
1322
+ "epoch": 1.0490842490842491,
1323
+ "grad_norm": 0.16661523282527924,
1324
+ "learning_rate": 9.507620164126613e-05,
1325
+ "loss": 0.0212,
1326
+ "step": 895
1327
+ },
1328
+ {
1329
+ "epoch": 1.054945054945055,
1330
+ "grad_norm": 0.08473166078329086,
1331
+ "learning_rate": 9.449003516998827e-05,
1332
+ "loss": 0.0285,
1333
+ "step": 900
1334
+ },
1335
+ {
1336
+ "epoch": 1.054945054945055,
1337
+ "eval_loss": 0.018958253785967827,
1338
+ "eval_runtime": 4.8179,
1339
+ "eval_samples_per_second": 5.812,
1340
+ "eval_steps_per_second": 2.906,
1341
+ "step": 900
1342
+ },
1343
+ {
1344
+ "epoch": 1.0608058608058608,
1345
+ "grad_norm": 0.04149739816784859,
1346
+ "learning_rate": 9.390386869871044e-05,
1347
+ "loss": 0.0138,
1348
+ "step": 905
1349
+ },
1350
+ {
1351
+ "epoch": 1.0666666666666667,
1352
+ "grad_norm": 0.05616866052150726,
1353
+ "learning_rate": 9.331770222743259e-05,
1354
+ "loss": 0.0194,
1355
+ "step": 910
1356
+ },
1357
+ {
1358
+ "epoch": 1.0725274725274725,
1359
+ "grad_norm": 0.16730394959449768,
1360
+ "learning_rate": 9.273153575615475e-05,
1361
+ "loss": 0.0359,
1362
+ "step": 915
1363
+ },
1364
+ {
1365
+ "epoch": 1.0783882783882783,
1366
+ "grad_norm": 0.06336849927902222,
1367
+ "learning_rate": 9.214536928487691e-05,
1368
+ "loss": 0.0147,
1369
+ "step": 920
1370
+ },
1371
+ {
1372
+ "epoch": 1.0842490842490842,
1373
+ "grad_norm": 0.0882687047123909,
1374
+ "learning_rate": 9.155920281359907e-05,
1375
+ "loss": 0.015,
1376
+ "step": 925
1377
+ },
1378
+ {
1379
+ "epoch": 1.09010989010989,
1380
+ "grad_norm": 0.05476200208067894,
1381
+ "learning_rate": 9.097303634232123e-05,
1382
+ "loss": 0.019,
1383
+ "step": 930
1384
+ },
1385
+ {
1386
+ "epoch": 1.095970695970696,
1387
+ "grad_norm": 0.05358652025461197,
1388
+ "learning_rate": 9.038686987104337e-05,
1389
+ "loss": 0.0174,
1390
+ "step": 935
1391
+ },
1392
+ {
1393
+ "epoch": 1.101831501831502,
1394
+ "grad_norm": 0.24095569550991058,
1395
+ "learning_rate": 8.980070339976554e-05,
1396
+ "loss": 0.0293,
1397
+ "step": 940
1398
+ },
1399
+ {
1400
+ "epoch": 1.1076923076923078,
1401
+ "grad_norm": 0.06653840094804764,
1402
+ "learning_rate": 8.921453692848769e-05,
1403
+ "loss": 0.0133,
1404
+ "step": 945
1405
+ },
1406
+ {
1407
+ "epoch": 1.1135531135531136,
1408
+ "grad_norm": 0.08839567750692368,
1409
+ "learning_rate": 8.862837045720985e-05,
1410
+ "loss": 0.0208,
1411
+ "step": 950
1412
+ },
1413
+ {
1414
+ "epoch": 1.1194139194139194,
1415
+ "grad_norm": 0.032916922122240067,
1416
+ "learning_rate": 8.804220398593201e-05,
1417
+ "loss": 0.0267,
1418
+ "step": 955
1419
+ },
1420
+ {
1421
+ "epoch": 1.1252747252747253,
1422
+ "grad_norm": 0.11819420754909515,
1423
+ "learning_rate": 8.745603751465417e-05,
1424
+ "loss": 0.0384,
1425
+ "step": 960
1426
+ },
1427
+ {
1428
+ "epoch": 1.1311355311355311,
1429
+ "grad_norm": 0.06757565587759018,
1430
+ "learning_rate": 8.686987104337633e-05,
1431
+ "loss": 0.0135,
1432
+ "step": 965
1433
+ },
1434
+ {
1435
+ "epoch": 1.136996336996337,
1436
+ "grad_norm": 0.0970228835940361,
1437
+ "learning_rate": 8.628370457209847e-05,
1438
+ "loss": 0.0166,
1439
+ "step": 970
1440
+ },
1441
+ {
1442
+ "epoch": 1.1428571428571428,
1443
+ "grad_norm": 0.13436350226402283,
1444
+ "learning_rate": 8.569753810082065e-05,
1445
+ "loss": 0.016,
1446
+ "step": 975
1447
+ },
1448
+ {
1449
+ "epoch": 1.1487179487179486,
1450
+ "grad_norm": 0.10584839433431625,
1451
+ "learning_rate": 8.511137162954279e-05,
1452
+ "loss": 0.0177,
1453
+ "step": 980
1454
+ },
1455
+ {
1456
+ "epoch": 1.1545787545787545,
1457
+ "grad_norm": 0.21206024289131165,
1458
+ "learning_rate": 8.452520515826495e-05,
1459
+ "loss": 0.0325,
1460
+ "step": 985
1461
+ },
1462
+ {
1463
+ "epoch": 1.1604395604395605,
1464
+ "grad_norm": 0.04815613850951195,
1465
+ "learning_rate": 8.393903868698711e-05,
1466
+ "loss": 0.0137,
1467
+ "step": 990
1468
+ },
1469
+ {
1470
+ "epoch": 1.1663003663003664,
1471
+ "grad_norm": 0.07466138154268265,
1472
+ "learning_rate": 8.335287221570927e-05,
1473
+ "loss": 0.0173,
1474
+ "step": 995
1475
+ },
1476
+ {
1477
+ "epoch": 1.1721611721611722,
1478
+ "grad_norm": 0.09366811066865921,
1479
+ "learning_rate": 8.276670574443143e-05,
1480
+ "loss": 0.025,
1481
+ "step": 1000
1482
+ },
1483
+ {
1484
+ "epoch": 1.1721611721611722,
1485
+ "eval_loss": 0.019907595589756966,
1486
+ "eval_runtime": 4.819,
1487
+ "eval_samples_per_second": 5.81,
1488
+ "eval_steps_per_second": 2.905,
1489
+ "step": 1000
1490
+ },
1491
+ {
1492
+ "epoch": 1.178021978021978,
1493
+ "grad_norm": 0.08527784794569016,
1494
+ "learning_rate": 8.218053927315357e-05,
1495
+ "loss": 0.0208,
1496
+ "step": 1005
1497
+ },
1498
+ {
1499
+ "epoch": 1.183882783882784,
1500
+ "grad_norm": 0.08328138291835785,
1501
+ "learning_rate": 8.159437280187575e-05,
1502
+ "loss": 0.0216,
1503
+ "step": 1010
1504
+ },
1505
+ {
1506
+ "epoch": 1.1897435897435897,
1507
+ "grad_norm": 0.05000188946723938,
1508
+ "learning_rate": 8.100820633059789e-05,
1509
+ "loss": 0.0211,
1510
+ "step": 1015
1511
+ },
1512
+ {
1513
+ "epoch": 1.1956043956043956,
1514
+ "grad_norm": 0.028807902708649635,
1515
+ "learning_rate": 8.042203985932005e-05,
1516
+ "loss": 0.0096,
1517
+ "step": 1020
1518
+ },
1519
+ {
1520
+ "epoch": 1.2014652014652014,
1521
+ "grad_norm": 0.20507606863975525,
1522
+ "learning_rate": 7.983587338804221e-05,
1523
+ "loss": 0.0222,
1524
+ "step": 1025
1525
+ },
1526
+ {
1527
+ "epoch": 1.2073260073260073,
1528
+ "grad_norm": 0.04885656014084816,
1529
+ "learning_rate": 7.924970691676437e-05,
1530
+ "loss": 0.0215,
1531
+ "step": 1030
1532
+ },
1533
+ {
1534
+ "epoch": 1.213186813186813,
1535
+ "grad_norm": 0.047489382326602936,
1536
+ "learning_rate": 7.866354044548652e-05,
1537
+ "loss": 0.0178,
1538
+ "step": 1035
1539
+ },
1540
+ {
1541
+ "epoch": 1.2190476190476192,
1542
+ "grad_norm": 0.05971779301762581,
1543
+ "learning_rate": 7.807737397420867e-05,
1544
+ "loss": 0.0176,
1545
+ "step": 1040
1546
+ },
1547
+ {
1548
+ "epoch": 1.224908424908425,
1549
+ "grad_norm": 0.04695736616849899,
1550
+ "learning_rate": 7.749120750293083e-05,
1551
+ "loss": 0.0148,
1552
+ "step": 1045
1553
+ },
1554
+ {
1555
+ "epoch": 1.2307692307692308,
1556
+ "grad_norm": 0.08131909370422363,
1557
+ "learning_rate": 7.690504103165299e-05,
1558
+ "loss": 0.0201,
1559
+ "step": 1050
1560
+ },
1561
+ {
1562
+ "epoch": 1.2366300366300367,
1563
+ "grad_norm": 0.06954577565193176,
1564
+ "learning_rate": 7.631887456037515e-05,
1565
+ "loss": 0.0149,
1566
+ "step": 1055
1567
+ },
1568
+ {
1569
+ "epoch": 1.2424908424908425,
1570
+ "grad_norm": 0.054430391639471054,
1571
+ "learning_rate": 7.57327080890973e-05,
1572
+ "loss": 0.0078,
1573
+ "step": 1060
1574
+ },
1575
+ {
1576
+ "epoch": 1.2483516483516484,
1577
+ "grad_norm": 0.12231959402561188,
1578
+ "learning_rate": 7.514654161781947e-05,
1579
+ "loss": 0.0258,
1580
+ "step": 1065
1581
+ },
1582
+ {
1583
+ "epoch": 1.2542124542124542,
1584
+ "grad_norm": 0.04983118548989296,
1585
+ "learning_rate": 7.456037514654162e-05,
1586
+ "loss": 0.0189,
1587
+ "step": 1070
1588
+ },
1589
+ {
1590
+ "epoch": 1.26007326007326,
1591
+ "grad_norm": 0.11981873214244843,
1592
+ "learning_rate": 7.397420867526378e-05,
1593
+ "loss": 0.0156,
1594
+ "step": 1075
1595
+ },
1596
+ {
1597
+ "epoch": 1.2659340659340659,
1598
+ "grad_norm": 0.03829724341630936,
1599
+ "learning_rate": 7.338804220398593e-05,
1600
+ "loss": 0.0162,
1601
+ "step": 1080
1602
+ },
1603
+ {
1604
+ "epoch": 1.2717948717948717,
1605
+ "grad_norm": 0.1572490632534027,
1606
+ "learning_rate": 7.28018757327081e-05,
1607
+ "loss": 0.0188,
1608
+ "step": 1085
1609
+ },
1610
+ {
1611
+ "epoch": 1.2776556776556776,
1612
+ "grad_norm": 0.122514508664608,
1613
+ "learning_rate": 7.221570926143025e-05,
1614
+ "loss": 0.0229,
1615
+ "step": 1090
1616
+ },
1617
+ {
1618
+ "epoch": 1.2835164835164834,
1619
+ "grad_norm": 0.06537042558193207,
1620
+ "learning_rate": 7.16295427901524e-05,
1621
+ "loss": 0.0222,
1622
+ "step": 1095
1623
+ },
1624
+ {
1625
+ "epoch": 1.2893772893772895,
1626
+ "grad_norm": 0.1269371509552002,
1627
+ "learning_rate": 7.104337631887457e-05,
1628
+ "loss": 0.0272,
1629
+ "step": 1100
1630
+ },
1631
+ {
1632
+ "epoch": 1.2893772893772895,
1633
+ "eval_loss": 0.015397748909890652,
1634
+ "eval_runtime": 4.7526,
1635
+ "eval_samples_per_second": 5.892,
1636
+ "eval_steps_per_second": 2.946,
1637
+ "step": 1100
1638
+ },
1639
+ {
1640
+ "epoch": 1.2952380952380953,
1641
+ "grad_norm": 0.07507819682359695,
1642
+ "learning_rate": 7.045720984759672e-05,
1643
+ "loss": 0.0256,
1644
+ "step": 1105
1645
+ },
1646
+ {
1647
+ "epoch": 1.3010989010989011,
1648
+ "grad_norm": 0.04198193550109863,
1649
+ "learning_rate": 6.987104337631888e-05,
1650
+ "loss": 0.0102,
1651
+ "step": 1110
1652
+ },
1653
+ {
1654
+ "epoch": 1.306959706959707,
1655
+ "grad_norm": 0.053751297295093536,
1656
+ "learning_rate": 6.928487690504104e-05,
1657
+ "loss": 0.0141,
1658
+ "step": 1115
1659
+ },
1660
+ {
1661
+ "epoch": 1.3128205128205128,
1662
+ "grad_norm": 0.12076237797737122,
1663
+ "learning_rate": 6.86987104337632e-05,
1664
+ "loss": 0.0165,
1665
+ "step": 1120
1666
+ },
1667
+ {
1668
+ "epoch": 1.3186813186813187,
1669
+ "grad_norm": 0.0769004300236702,
1670
+ "learning_rate": 6.811254396248535e-05,
1671
+ "loss": 0.0191,
1672
+ "step": 1125
1673
+ },
1674
+ {
1675
+ "epoch": 1.3245421245421245,
1676
+ "grad_norm": 0.08021704852581024,
1677
+ "learning_rate": 6.75263774912075e-05,
1678
+ "loss": 0.0253,
1679
+ "step": 1130
1680
+ },
1681
+ {
1682
+ "epoch": 1.3304029304029303,
1683
+ "grad_norm": 0.09786754846572876,
1684
+ "learning_rate": 6.694021101992967e-05,
1685
+ "loss": 0.0191,
1686
+ "step": 1135
1687
+ },
1688
+ {
1689
+ "epoch": 1.3362637362637364,
1690
+ "grad_norm": 0.06878714263439178,
1691
+ "learning_rate": 6.635404454865182e-05,
1692
+ "loss": 0.0326,
1693
+ "step": 1140
1694
+ },
1695
+ {
1696
+ "epoch": 1.3421245421245422,
1697
+ "grad_norm": 0.11297193914651871,
1698
+ "learning_rate": 6.576787807737398e-05,
1699
+ "loss": 0.0185,
1700
+ "step": 1145
1701
+ },
1702
+ {
1703
+ "epoch": 1.347985347985348,
1704
+ "grad_norm": 0.10731342434883118,
1705
+ "learning_rate": 6.518171160609614e-05,
1706
+ "loss": 0.0168,
1707
+ "step": 1150
1708
+ },
1709
+ {
1710
+ "epoch": 1.353846153846154,
1711
+ "grad_norm": 0.08888328820466995,
1712
+ "learning_rate": 6.45955451348183e-05,
1713
+ "loss": 0.0182,
1714
+ "step": 1155
1715
+ },
1716
+ {
1717
+ "epoch": 1.3597069597069598,
1718
+ "grad_norm": 0.1666301190853119,
1719
+ "learning_rate": 6.400937866354045e-05,
1720
+ "loss": 0.0254,
1721
+ "step": 1160
1722
+ },
1723
+ {
1724
+ "epoch": 1.3655677655677656,
1725
+ "grad_norm": 0.1334419697523117,
1726
+ "learning_rate": 6.34232121922626e-05,
1727
+ "loss": 0.0215,
1728
+ "step": 1165
1729
+ },
1730
+ {
1731
+ "epoch": 1.3714285714285714,
1732
+ "grad_norm": 0.05616243556141853,
1733
+ "learning_rate": 6.283704572098477e-05,
1734
+ "loss": 0.0156,
1735
+ "step": 1170
1736
+ },
1737
+ {
1738
+ "epoch": 1.3772893772893773,
1739
+ "grad_norm": 0.1660885214805603,
1740
+ "learning_rate": 6.225087924970692e-05,
1741
+ "loss": 0.0241,
1742
+ "step": 1175
1743
+ },
1744
+ {
1745
+ "epoch": 1.3831501831501831,
1746
+ "grad_norm": 0.09245380759239197,
1747
+ "learning_rate": 6.166471277842908e-05,
1748
+ "loss": 0.0222,
1749
+ "step": 1180
1750
+ },
1751
+ {
1752
+ "epoch": 1.389010989010989,
1753
+ "grad_norm": 0.08635041117668152,
1754
+ "learning_rate": 6.107854630715122e-05,
1755
+ "loss": 0.0203,
1756
+ "step": 1185
1757
+ },
1758
+ {
1759
+ "epoch": 1.3948717948717948,
1760
+ "grad_norm": 0.07752135396003723,
1761
+ "learning_rate": 6.049237983587339e-05,
1762
+ "loss": 0.0216,
1763
+ "step": 1190
1764
+ },
1765
+ {
1766
+ "epoch": 1.4007326007326006,
1767
+ "grad_norm": 0.10603225976228714,
1768
+ "learning_rate": 5.990621336459554e-05,
1769
+ "loss": 0.0205,
1770
+ "step": 1195
1771
+ },
1772
+ {
1773
+ "epoch": 1.4065934065934065,
1774
+ "grad_norm": 0.04343140870332718,
1775
+ "learning_rate": 5.932004689331771e-05,
1776
+ "loss": 0.0105,
1777
+ "step": 1200
1778
+ },
1779
+ {
1780
+ "epoch": 1.4065934065934065,
1781
+ "eval_loss": 0.015952473506331444,
1782
+ "eval_runtime": 4.7686,
1783
+ "eval_samples_per_second": 5.872,
1784
+ "eval_steps_per_second": 2.936,
1785
+ "step": 1200
1786
+ },
1787
+ {
1788
+ "epoch": 1.4124542124542123,
1789
+ "grad_norm": 0.12541887164115906,
1790
+ "learning_rate": 5.873388042203987e-05,
1791
+ "loss": 0.0217,
1792
+ "step": 1205
1793
+ },
1794
+ {
1795
+ "epoch": 1.4183150183150184,
1796
+ "grad_norm": 0.05417335778474808,
1797
+ "learning_rate": 5.814771395076202e-05,
1798
+ "loss": 0.0205,
1799
+ "step": 1210
1800
+ },
1801
+ {
1802
+ "epoch": 1.4241758241758242,
1803
+ "grad_norm": 0.04461506009101868,
1804
+ "learning_rate": 5.756154747948418e-05,
1805
+ "loss": 0.0211,
1806
+ "step": 1215
1807
+ },
1808
+ {
1809
+ "epoch": 1.43003663003663,
1810
+ "grad_norm": 0.09423286467790604,
1811
+ "learning_rate": 5.697538100820633e-05,
1812
+ "loss": 0.0238,
1813
+ "step": 1220
1814
+ },
1815
+ {
1816
+ "epoch": 1.435897435897436,
1817
+ "grad_norm": 0.050094570964574814,
1818
+ "learning_rate": 5.638921453692849e-05,
1819
+ "loss": 0.0163,
1820
+ "step": 1225
1821
+ },
1822
+ {
1823
+ "epoch": 1.4417582417582417,
1824
+ "grad_norm": 0.13104532659053802,
1825
+ "learning_rate": 5.580304806565064e-05,
1826
+ "loss": 0.0219,
1827
+ "step": 1230
1828
+ },
1829
+ {
1830
+ "epoch": 1.4476190476190476,
1831
+ "grad_norm": 0.08238503336906433,
1832
+ "learning_rate": 5.52168815943728e-05,
1833
+ "loss": 0.0126,
1834
+ "step": 1235
1835
+ },
1836
+ {
1837
+ "epoch": 1.4534798534798534,
1838
+ "grad_norm": 0.1029452383518219,
1839
+ "learning_rate": 5.463071512309497e-05,
1840
+ "loss": 0.0247,
1841
+ "step": 1240
1842
+ },
1843
+ {
1844
+ "epoch": 1.4593406593406593,
1845
+ "grad_norm": 0.05564792454242706,
1846
+ "learning_rate": 5.404454865181712e-05,
1847
+ "loss": 0.0212,
1848
+ "step": 1245
1849
+ },
1850
+ {
1851
+ "epoch": 1.4652014652014653,
1852
+ "grad_norm": 0.08589282631874084,
1853
+ "learning_rate": 5.345838218053928e-05,
1854
+ "loss": 0.0184,
1855
+ "step": 1250
1856
+ },
1857
+ {
1858
+ "epoch": 1.4710622710622712,
1859
+ "grad_norm": 0.15644195675849915,
1860
+ "learning_rate": 5.287221570926143e-05,
1861
+ "loss": 0.0165,
1862
+ "step": 1255
1863
+ },
1864
+ {
1865
+ "epoch": 1.476923076923077,
1866
+ "grad_norm": 0.11301274597644806,
1867
+ "learning_rate": 5.228604923798359e-05,
1868
+ "loss": 0.0322,
1869
+ "step": 1260
1870
+ },
1871
+ {
1872
+ "epoch": 1.4827838827838828,
1873
+ "grad_norm": 0.044553741812705994,
1874
+ "learning_rate": 5.1699882766705743e-05,
1875
+ "loss": 0.0183,
1876
+ "step": 1265
1877
+ },
1878
+ {
1879
+ "epoch": 1.4886446886446887,
1880
+ "grad_norm": 0.06141185760498047,
1881
+ "learning_rate": 5.11137162954279e-05,
1882
+ "loss": 0.0102,
1883
+ "step": 1270
1884
+ },
1885
+ {
1886
+ "epoch": 1.4945054945054945,
1887
+ "grad_norm": 0.08107537031173706,
1888
+ "learning_rate": 5.0527549824150055e-05,
1889
+ "loss": 0.0215,
1890
+ "step": 1275
1891
+ },
1892
+ {
1893
+ "epoch": 1.5003663003663004,
1894
+ "grad_norm": 0.06017793342471123,
1895
+ "learning_rate": 4.9941383352872214e-05,
1896
+ "loss": 0.0176,
1897
+ "step": 1280
1898
+ },
1899
+ {
1900
+ "epoch": 1.5062271062271062,
1901
+ "grad_norm": 0.051033902913331985,
1902
+ "learning_rate": 4.9355216881594373e-05,
1903
+ "loss": 0.0149,
1904
+ "step": 1285
1905
+ },
1906
+ {
1907
+ "epoch": 1.512087912087912,
1908
+ "grad_norm": 0.124452143907547,
1909
+ "learning_rate": 4.876905041031653e-05,
1910
+ "loss": 0.0209,
1911
+ "step": 1290
1912
+ },
1913
+ {
1914
+ "epoch": 1.5179487179487179,
1915
+ "grad_norm": 0.1616523712873459,
1916
+ "learning_rate": 4.8182883939038685e-05,
1917
+ "loss": 0.0192,
1918
+ "step": 1295
1919
+ },
1920
+ {
1921
+ "epoch": 1.5238095238095237,
1922
+ "grad_norm": 0.07067764550447464,
1923
+ "learning_rate": 4.759671746776085e-05,
1924
+ "loss": 0.0279,
1925
+ "step": 1300
1926
+ },
1927
+ {
1928
+ "epoch": 1.5238095238095237,
1929
+ "eval_loss": 0.017229218035936356,
1930
+ "eval_runtime": 4.6702,
1931
+ "eval_samples_per_second": 5.996,
1932
+ "eval_steps_per_second": 2.998,
1933
+ "step": 1300
1934
+ },
1935
+ {
1936
+ "epoch": 1.5296703296703296,
1937
+ "grad_norm": 0.06332267820835114,
1938
+ "learning_rate": 4.7010550996483003e-05,
1939
+ "loss": 0.0253,
1940
+ "step": 1305
1941
+ },
1942
+ {
1943
+ "epoch": 1.5355311355311354,
1944
+ "grad_norm": 0.08032066375017166,
1945
+ "learning_rate": 4.642438452520516e-05,
1946
+ "loss": 0.0128,
1947
+ "step": 1310
1948
+ },
1949
+ {
1950
+ "epoch": 1.5413919413919412,
1951
+ "grad_norm": 0.11456907540559769,
1952
+ "learning_rate": 4.5838218053927315e-05,
1953
+ "loss": 0.0127,
1954
+ "step": 1315
1955
+ },
1956
+ {
1957
+ "epoch": 1.5472527472527473,
1958
+ "grad_norm": 0.19653138518333435,
1959
+ "learning_rate": 4.5252051582649474e-05,
1960
+ "loss": 0.0236,
1961
+ "step": 1320
1962
+ },
1963
+ {
1964
+ "epoch": 1.5531135531135531,
1965
+ "grad_norm": 0.08195839822292328,
1966
+ "learning_rate": 4.4665885111371633e-05,
1967
+ "loss": 0.0173,
1968
+ "step": 1325
1969
+ },
1970
+ {
1971
+ "epoch": 1.558974358974359,
1972
+ "grad_norm": 0.11376089602708817,
1973
+ "learning_rate": 4.4079718640093786e-05,
1974
+ "loss": 0.0115,
1975
+ "step": 1330
1976
+ },
1977
+ {
1978
+ "epoch": 1.5648351648351648,
1979
+ "grad_norm": 0.055264201015233994,
1980
+ "learning_rate": 4.3493552168815945e-05,
1981
+ "loss": 0.0198,
1982
+ "step": 1335
1983
+ },
1984
+ {
1985
+ "epoch": 1.5706959706959707,
1986
+ "grad_norm": 0.13038881123065948,
1987
+ "learning_rate": 4.2907385697538104e-05,
1988
+ "loss": 0.0231,
1989
+ "step": 1340
1990
+ },
1991
+ {
1992
+ "epoch": 1.5765567765567765,
1993
+ "grad_norm": 0.0317939892411232,
1994
+ "learning_rate": 4.2321219226260263e-05,
1995
+ "loss": 0.0083,
1996
+ "step": 1345
1997
+ },
1998
+ {
1999
+ "epoch": 1.5824175824175826,
2000
+ "grad_norm": 0.151336207985878,
2001
+ "learning_rate": 4.1735052754982416e-05,
2002
+ "loss": 0.0225,
2003
+ "step": 1350
2004
+ },
2005
+ {
2006
+ "epoch": 1.5882783882783884,
2007
+ "grad_norm": 0.07817093282938004,
2008
+ "learning_rate": 4.1148886283704575e-05,
2009
+ "loss": 0.0226,
2010
+ "step": 1355
2011
+ },
2012
+ {
2013
+ "epoch": 1.5941391941391942,
2014
+ "grad_norm": 0.1341279149055481,
2015
+ "learning_rate": 4.056271981242673e-05,
2016
+ "loss": 0.0263,
2017
+ "step": 1360
2018
+ },
2019
+ {
2020
+ "epoch": 1.6,
2021
+ "grad_norm": 0.06353727728128433,
2022
+ "learning_rate": 3.997655334114889e-05,
2023
+ "loss": 0.0198,
2024
+ "step": 1365
2025
+ },
2026
+ {
2027
+ "epoch": 1.605860805860806,
2028
+ "grad_norm": 0.11177901178598404,
2029
+ "learning_rate": 3.9390386869871046e-05,
2030
+ "loss": 0.0172,
2031
+ "step": 1370
2032
+ },
2033
+ {
2034
+ "epoch": 1.6117216117216118,
2035
+ "grad_norm": 0.047024596482515335,
2036
+ "learning_rate": 3.88042203985932e-05,
2037
+ "loss": 0.0207,
2038
+ "step": 1375
2039
+ },
2040
+ {
2041
+ "epoch": 1.6175824175824176,
2042
+ "grad_norm": 0.04343528300523758,
2043
+ "learning_rate": 3.8218053927315364e-05,
2044
+ "loss": 0.0214,
2045
+ "step": 1380
2046
+ },
2047
+ {
2048
+ "epoch": 1.6234432234432234,
2049
+ "grad_norm": 0.08330193161964417,
2050
+ "learning_rate": 3.763188745603752e-05,
2051
+ "loss": 0.0286,
2052
+ "step": 1385
2053
+ },
2054
+ {
2055
+ "epoch": 1.6293040293040293,
2056
+ "grad_norm": 0.0811009556055069,
2057
+ "learning_rate": 3.7045720984759676e-05,
2058
+ "loss": 0.0148,
2059
+ "step": 1390
2060
+ },
2061
+ {
2062
+ "epoch": 1.6351648351648351,
2063
+ "grad_norm": 0.1049441322684288,
2064
+ "learning_rate": 3.645955451348183e-05,
2065
+ "loss": 0.0184,
2066
+ "step": 1395
2067
+ },
2068
+ {
2069
+ "epoch": 1.641025641025641,
2070
+ "grad_norm": 0.11944428086280823,
2071
+ "learning_rate": 3.587338804220399e-05,
2072
+ "loss": 0.0122,
2073
+ "step": 1400
2074
+ },
2075
+ {
2076
+ "epoch": 1.641025641025641,
2077
+ "eval_loss": 0.017561230808496475,
2078
+ "eval_runtime": 4.685,
2079
+ "eval_samples_per_second": 5.977,
2080
+ "eval_steps_per_second": 2.988,
2081
+ "step": 1400
2082
+ },
2083
+ {
2084
+ "epoch": 1.6468864468864468,
2085
+ "grad_norm": 0.14023366570472717,
2086
+ "learning_rate": 3.528722157092615e-05,
2087
+ "loss": 0.0178,
2088
+ "step": 1405
2089
+ },
2090
+ {
2091
+ "epoch": 1.6527472527472526,
2092
+ "grad_norm": 0.14057691395282745,
2093
+ "learning_rate": 3.47010550996483e-05,
2094
+ "loss": 0.0268,
2095
+ "step": 1410
2096
+ },
2097
+ {
2098
+ "epoch": 1.6586080586080585,
2099
+ "grad_norm": 0.1253061443567276,
2100
+ "learning_rate": 3.411488862837046e-05,
2101
+ "loss": 0.0266,
2102
+ "step": 1415
2103
+ },
2104
+ {
2105
+ "epoch": 1.6644688644688643,
2106
+ "grad_norm": 0.03431854769587517,
2107
+ "learning_rate": 3.352872215709262e-05,
2108
+ "loss": 0.02,
2109
+ "step": 1420
2110
+ },
2111
+ {
2112
+ "epoch": 1.6703296703296702,
2113
+ "grad_norm": 0.13929079473018646,
2114
+ "learning_rate": 3.294255568581478e-05,
2115
+ "loss": 0.0226,
2116
+ "step": 1425
2117
+ },
2118
+ {
2119
+ "epoch": 1.6761904761904762,
2120
+ "grad_norm": 0.06429693102836609,
2121
+ "learning_rate": 3.235638921453693e-05,
2122
+ "loss": 0.0225,
2123
+ "step": 1430
2124
+ },
2125
+ {
2126
+ "epoch": 1.682051282051282,
2127
+ "grad_norm": 0.029311953112483025,
2128
+ "learning_rate": 3.177022274325909e-05,
2129
+ "loss": 0.0161,
2130
+ "step": 1435
2131
+ },
2132
+ {
2133
+ "epoch": 1.687912087912088,
2134
+ "grad_norm": 0.04346455633640289,
2135
+ "learning_rate": 3.118405627198124e-05,
2136
+ "loss": 0.0155,
2137
+ "step": 1440
2138
+ },
2139
+ {
2140
+ "epoch": 1.6937728937728938,
2141
+ "grad_norm": 0.09009824693202972,
2142
+ "learning_rate": 3.05978898007034e-05,
2143
+ "loss": 0.0153,
2144
+ "step": 1445
2145
+ },
2146
+ {
2147
+ "epoch": 1.6996336996336996,
2148
+ "grad_norm": 0.071926549077034,
2149
+ "learning_rate": 3.0011723329425556e-05,
2150
+ "loss": 0.0136,
2151
+ "step": 1450
2152
+ },
2153
+ {
2154
+ "epoch": 1.7054945054945057,
2155
+ "grad_norm": 0.06461833417415619,
2156
+ "learning_rate": 2.9425556858147718e-05,
2157
+ "loss": 0.0237,
2158
+ "step": 1455
2159
+ },
2160
+ {
2161
+ "epoch": 1.7113553113553115,
2162
+ "grad_norm": 0.039929524064064026,
2163
+ "learning_rate": 2.8839390386869874e-05,
2164
+ "loss": 0.0187,
2165
+ "step": 1460
2166
+ },
2167
+ {
2168
+ "epoch": 1.7172161172161173,
2169
+ "grad_norm": 0.0534372515976429,
2170
+ "learning_rate": 2.825322391559203e-05,
2171
+ "loss": 0.0192,
2172
+ "step": 1465
2173
+ },
2174
+ {
2175
+ "epoch": 1.7230769230769232,
2176
+ "grad_norm": 0.1467376947402954,
2177
+ "learning_rate": 2.7667057444314186e-05,
2178
+ "loss": 0.0203,
2179
+ "step": 1470
2180
+ },
2181
+ {
2182
+ "epoch": 1.728937728937729,
2183
+ "grad_norm": 0.0830003172159195,
2184
+ "learning_rate": 2.7080890973036345e-05,
2185
+ "loss": 0.0188,
2186
+ "step": 1475
2187
+ },
2188
+ {
2189
+ "epoch": 1.7347985347985349,
2190
+ "grad_norm": 0.07220768928527832,
2191
+ "learning_rate": 2.64947245017585e-05,
2192
+ "loss": 0.0118,
2193
+ "step": 1480
2194
+ },
2195
+ {
2196
+ "epoch": 1.7406593406593407,
2197
+ "grad_norm": 0.0751115009188652,
2198
+ "learning_rate": 2.5908558030480656e-05,
2199
+ "loss": 0.0156,
2200
+ "step": 1485
2201
+ },
2202
+ {
2203
+ "epoch": 1.7465201465201465,
2204
+ "grad_norm": 0.07690921425819397,
2205
+ "learning_rate": 2.5322391559202812e-05,
2206
+ "loss": 0.0347,
2207
+ "step": 1490
2208
+ },
2209
+ {
2210
+ "epoch": 1.7523809523809524,
2211
+ "grad_norm": 0.05416159704327583,
2212
+ "learning_rate": 2.473622508792497e-05,
2213
+ "loss": 0.0167,
2214
+ "step": 1495
2215
+ },
2216
+ {
2217
+ "epoch": 1.7582417582417582,
2218
+ "grad_norm": 0.0676250010728836,
2219
+ "learning_rate": 2.4150058616647127e-05,
2220
+ "loss": 0.0205,
2221
+ "step": 1500
2222
+ },
2223
+ {
2224
+ "epoch": 1.7582417582417582,
2225
+ "eval_loss": 0.019522378221154213,
2226
+ "eval_runtime": 4.6942,
2227
+ "eval_samples_per_second": 5.965,
2228
+ "eval_steps_per_second": 2.982,
2229
+ "step": 1500
2230
+ },
2231
+ {
2232
+ "epoch": 1.764102564102564,
2233
+ "grad_norm": 0.08909754455089569,
2234
+ "learning_rate": 2.3563892145369286e-05,
2235
+ "loss": 0.0249,
2236
+ "step": 1505
2237
+ },
2238
+ {
2239
+ "epoch": 1.76996336996337,
2240
+ "grad_norm": 0.042161233723163605,
2241
+ "learning_rate": 2.2977725674091442e-05,
2242
+ "loss": 0.0117,
2243
+ "step": 1510
2244
+ },
2245
+ {
2246
+ "epoch": 1.7758241758241757,
2247
+ "grad_norm": 0.07136218249797821,
2248
+ "learning_rate": 2.23915592028136e-05,
2249
+ "loss": 0.0212,
2250
+ "step": 1515
2251
+ },
2252
+ {
2253
+ "epoch": 1.7816849816849816,
2254
+ "grad_norm": 0.14128735661506653,
2255
+ "learning_rate": 2.1805392731535757e-05,
2256
+ "loss": 0.0189,
2257
+ "step": 1520
2258
+ },
2259
+ {
2260
+ "epoch": 1.7875457875457874,
2261
+ "grad_norm": 0.05959760770201683,
2262
+ "learning_rate": 2.1219226260257916e-05,
2263
+ "loss": 0.0119,
2264
+ "step": 1525
2265
+ },
2266
+ {
2267
+ "epoch": 1.7934065934065933,
2268
+ "grad_norm": 0.038479190319776535,
2269
+ "learning_rate": 2.0633059788980072e-05,
2270
+ "loss": 0.013,
2271
+ "step": 1530
2272
+ },
2273
+ {
2274
+ "epoch": 1.7992673992673993,
2275
+ "grad_norm": 0.09512809664011002,
2276
+ "learning_rate": 2.0046893317702228e-05,
2277
+ "loss": 0.0148,
2278
+ "step": 1535
2279
+ },
2280
+ {
2281
+ "epoch": 1.8051282051282052,
2282
+ "grad_norm": 0.14848454296588898,
2283
+ "learning_rate": 1.9460726846424384e-05,
2284
+ "loss": 0.019,
2285
+ "step": 1540
2286
+ },
2287
+ {
2288
+ "epoch": 1.810989010989011,
2289
+ "grad_norm": 0.10240516811609268,
2290
+ "learning_rate": 1.8874560375146543e-05,
2291
+ "loss": 0.017,
2292
+ "step": 1545
2293
+ },
2294
+ {
2295
+ "epoch": 1.8168498168498168,
2296
+ "grad_norm": 0.09345954656600952,
2297
+ "learning_rate": 1.82883939038687e-05,
2298
+ "loss": 0.0237,
2299
+ "step": 1550
2300
+ },
2301
+ {
2302
+ "epoch": 1.8227106227106227,
2303
+ "grad_norm": 0.03815275430679321,
2304
+ "learning_rate": 1.7702227432590858e-05,
2305
+ "loss": 0.0188,
2306
+ "step": 1555
2307
+ },
2308
+ {
2309
+ "epoch": 1.8285714285714287,
2310
+ "grad_norm": 0.027827398851513863,
2311
+ "learning_rate": 1.7116060961313014e-05,
2312
+ "loss": 0.0183,
2313
+ "step": 1560
2314
+ },
2315
+ {
2316
+ "epoch": 1.8344322344322346,
2317
+ "grad_norm": 0.08811303228139877,
2318
+ "learning_rate": 1.6529894490035173e-05,
2319
+ "loss": 0.0191,
2320
+ "step": 1565
2321
+ },
2322
+ {
2323
+ "epoch": 1.8402930402930404,
2324
+ "grad_norm": 0.03119056299328804,
2325
+ "learning_rate": 1.594372801875733e-05,
2326
+ "loss": 0.0183,
2327
+ "step": 1570
2328
+ },
2329
+ {
2330
+ "epoch": 1.8461538461538463,
2331
+ "grad_norm": 0.09752997010946274,
2332
+ "learning_rate": 1.5357561547479485e-05,
2333
+ "loss": 0.0161,
2334
+ "step": 1575
2335
+ },
2336
+ {
2337
+ "epoch": 1.852014652014652,
2338
+ "grad_norm": 0.0855243131518364,
2339
+ "learning_rate": 1.477139507620164e-05,
2340
+ "loss": 0.015,
2341
+ "step": 1580
2342
+ },
2343
+ {
2344
+ "epoch": 1.857875457875458,
2345
+ "grad_norm": 0.08388842642307281,
2346
+ "learning_rate": 1.41852286049238e-05,
2347
+ "loss": 0.0148,
2348
+ "step": 1585
2349
+ },
2350
+ {
2351
+ "epoch": 1.8637362637362638,
2352
+ "grad_norm": 0.10147551447153091,
2353
+ "learning_rate": 1.3599062133645957e-05,
2354
+ "loss": 0.0154,
2355
+ "step": 1590
2356
+ },
2357
+ {
2358
+ "epoch": 1.8695970695970696,
2359
+ "grad_norm": 0.0457012839615345,
2360
+ "learning_rate": 1.3012895662368113e-05,
2361
+ "loss": 0.0186,
2362
+ "step": 1595
2363
+ },
2364
+ {
2365
+ "epoch": 1.8754578754578755,
2366
+ "grad_norm": 0.03654688224196434,
2367
+ "learning_rate": 1.242672919109027e-05,
2368
+ "loss": 0.0321,
2369
+ "step": 1600
2370
+ },
2371
+ {
2372
+ "epoch": 1.8754578754578755,
2373
+ "eval_loss": 0.017443044111132622,
2374
+ "eval_runtime": 4.7108,
2375
+ "eval_samples_per_second": 5.944,
2376
+ "eval_steps_per_second": 2.972,
2377
+ "step": 1600
2378
+ },
2379
+ {
2380
+ "epoch": 1.8813186813186813,
2381
+ "grad_norm": 0.07887323200702667,
2382
+ "learning_rate": 1.1840562719812428e-05,
2383
+ "loss": 0.0142,
2384
+ "step": 1605
2385
+ },
2386
+ {
2387
+ "epoch": 1.8871794871794871,
2388
+ "grad_norm": 0.11328335106372833,
2389
+ "learning_rate": 1.1254396248534585e-05,
2390
+ "loss": 0.025,
2391
+ "step": 1610
2392
+ },
2393
+ {
2394
+ "epoch": 1.893040293040293,
2395
+ "grad_norm": 0.09318089485168457,
2396
+ "learning_rate": 1.0668229777256741e-05,
2397
+ "loss": 0.0204,
2398
+ "step": 1615
2399
+ },
2400
+ {
2401
+ "epoch": 1.8989010989010988,
2402
+ "grad_norm": 0.06992164254188538,
2403
+ "learning_rate": 1.0082063305978899e-05,
2404
+ "loss": 0.0135,
2405
+ "step": 1620
2406
+ },
2407
+ {
2408
+ "epoch": 1.9047619047619047,
2409
+ "grad_norm": 0.14927181601524353,
2410
+ "learning_rate": 9.495896834701056e-06,
2411
+ "loss": 0.0245,
2412
+ "step": 1625
2413
+ },
2414
+ {
2415
+ "epoch": 1.9106227106227105,
2416
+ "grad_norm": 0.11408836394548416,
2417
+ "learning_rate": 8.909730363423214e-06,
2418
+ "loss": 0.0161,
2419
+ "step": 1630
2420
+ },
2421
+ {
2422
+ "epoch": 1.9164835164835163,
2423
+ "grad_norm": 0.06911155581474304,
2424
+ "learning_rate": 8.32356389214537e-06,
2425
+ "loss": 0.0154,
2426
+ "step": 1635
2427
+ },
2428
+ {
2429
+ "epoch": 1.9223443223443224,
2430
+ "grad_norm": 0.11621779948472977,
2431
+ "learning_rate": 7.737397420867527e-06,
2432
+ "loss": 0.0144,
2433
+ "step": 1640
2434
+ },
2435
+ {
2436
+ "epoch": 1.9282051282051282,
2437
+ "grad_norm": 0.046058397740125656,
2438
+ "learning_rate": 7.151230949589684e-06,
2439
+ "loss": 0.0093,
2440
+ "step": 1645
2441
+ },
2442
+ {
2443
+ "epoch": 1.934065934065934,
2444
+ "grad_norm": 0.11228576302528381,
2445
+ "learning_rate": 6.565064478311841e-06,
2446
+ "loss": 0.022,
2447
+ "step": 1650
2448
+ },
2449
+ {
2450
+ "epoch": 1.93992673992674,
2451
+ "grad_norm": 0.1315338909626007,
2452
+ "learning_rate": 5.978898007033998e-06,
2453
+ "loss": 0.0193,
2454
+ "step": 1655
2455
+ },
2456
+ {
2457
+ "epoch": 1.9457875457875458,
2458
+ "grad_norm": 0.040056392550468445,
2459
+ "learning_rate": 5.3927315357561546e-06,
2460
+ "loss": 0.0132,
2461
+ "step": 1660
2462
+ },
2463
+ {
2464
+ "epoch": 1.9516483516483516,
2465
+ "grad_norm": 0.10741738229990005,
2466
+ "learning_rate": 4.806565064478312e-06,
2467
+ "loss": 0.0352,
2468
+ "step": 1665
2469
+ },
2470
+ {
2471
+ "epoch": 1.9575091575091577,
2472
+ "grad_norm": 0.059029560536146164,
2473
+ "learning_rate": 4.220398593200469e-06,
2474
+ "loss": 0.019,
2475
+ "step": 1670
2476
+ },
2477
+ {
2478
+ "epoch": 1.9633699633699635,
2479
+ "grad_norm": 0.06789711117744446,
2480
+ "learning_rate": 3.6342321219226262e-06,
2481
+ "loss": 0.0189,
2482
+ "step": 1675
2483
+ },
2484
+ {
2485
+ "epoch": 1.9692307692307693,
2486
+ "grad_norm": 0.02918117679655552,
2487
+ "learning_rate": 3.0480656506447833e-06,
2488
+ "loss": 0.0259,
2489
+ "step": 1680
2490
+ },
2491
+ {
2492
+ "epoch": 1.9750915750915752,
2493
+ "grad_norm": 0.08073403686285019,
2494
+ "learning_rate": 2.4618991793669404e-06,
2495
+ "loss": 0.0286,
2496
+ "step": 1685
2497
+ },
2498
+ {
2499
+ "epoch": 1.980952380952381,
2500
+ "grad_norm": 0.1617717295885086,
2501
+ "learning_rate": 1.8757327080890972e-06,
2502
+ "loss": 0.0191,
2503
+ "step": 1690
2504
+ },
2505
+ {
2506
+ "epoch": 1.9868131868131869,
2507
+ "grad_norm": 0.06613462418317795,
2508
+ "learning_rate": 1.2895662368112545e-06,
2509
+ "loss": 0.0128,
2510
+ "step": 1695
2511
+ },
2512
+ {
2513
+ "epoch": 1.9926739926739927,
2514
+ "grad_norm": 0.08398256450891495,
2515
+ "learning_rate": 7.033997655334116e-07,
2516
+ "loss": 0.0118,
2517
+ "step": 1700
2518
+ },
2519
+ {
2520
+ "epoch": 1.9926739926739927,
2521
+ "eval_loss": 0.018166696652770042,
2522
+ "eval_runtime": 4.6959,
2523
+ "eval_samples_per_second": 5.963,
2524
+ "eval_steps_per_second": 2.981,
2525
+ "step": 1700
2526
+ },
2527
+ {
2528
+ "epoch": 1.9985347985347985,
2529
+ "grad_norm": 0.0773661658167839,
2530
+ "learning_rate": 1.1723329425556858e-07,
2531
+ "loss": 0.0146,
2532
+ "step": 1705
2533
+ }
2534
+ ],
2535
+ "logging_steps": 5,
2536
+ "max_steps": 1706,
2537
+ "num_input_tokens_seen": 0,
2538
+ "num_train_epochs": 2,
2539
+ "save_steps": 1000,
2540
+ "stateful_callbacks": {
2541
+ "TrainerControl": {
2542
+ "args": {
2543
+ "should_epoch_stop": false,
2544
+ "should_evaluate": false,
2545
+ "should_log": false,
2546
+ "should_save": true,
2547
+ "should_training_stop": true
2548
+ },
2549
+ "attributes": {}
2550
+ }
2551
+ },
2552
+ "total_flos": 5.764499308335456e+16,
2553
+ "train_batch_size": 2,
2554
+ "trial_name": null,
2555
+ "trial_params": null
2556
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a1417d5a273a397df996a9b4c569c3bc6e83978393843d305b311edfe15d0c4
3
+ size 5112