Daniel23Stack commited on
Commit
878a652
1 Parent(s): f175a7c

Upload 26 files

Browse files
Files changed (26) hide show
  1. aliceinwonderland/README.md +202 -0
  2. aliceinwonderland/adapter_config.json +27 -0
  3. aliceinwonderland/adapter_model.bin +3 -0
  4. aliceinwonderland/checkpoint-15-loss-1_17/README.md +202 -0
  5. aliceinwonderland/checkpoint-15-loss-1_17/adapter_config.json +27 -0
  6. aliceinwonderland/checkpoint-15-loss-1_17/adapter_model.bin +3 -0
  7. aliceinwonderland/checkpoint-15-loss-1_17/training_log.json +19 -0
  8. aliceinwonderland/checkpoint-15-loss-1_17/training_prompt.json +3 -0
  9. aliceinwonderland/checkpoint-19-loss-0_90/README.md +202 -0
  10. aliceinwonderland/checkpoint-19-loss-0_90/adapter_config.json +27 -0
  11. aliceinwonderland/checkpoint-19-loss-0_90/adapter_model.bin +3 -0
  12. aliceinwonderland/checkpoint-19-loss-0_90/training_log.json +19 -0
  13. aliceinwonderland/checkpoint-19-loss-0_90/training_prompt.json +3 -0
  14. aliceinwonderland/checkpoint-23-loss-0_60/README.md +202 -0
  15. aliceinwonderland/checkpoint-23-loss-0_60/adapter_config.json +27 -0
  16. aliceinwonderland/checkpoint-23-loss-0_60/adapter_model.bin +3 -0
  17. aliceinwonderland/checkpoint-23-loss-0_60/training_log.json +19 -0
  18. aliceinwonderland/checkpoint-23-loss-0_60/training_prompt.json +3 -0
  19. aliceinwonderland/runs/Jun04_00-27-53/events.out.tfevents.1717478875.DESKTOP-7QRHF82.5780.0 +3 -0
  20. aliceinwonderland/runs/Jun04_00-32-30/events.out.tfevents.1717479151.DESKTOP-7QRHF82.5780.1 +3 -0
  21. aliceinwonderland/runs/Jun04_00-34-12/events.out.tfevents.1717479252.DESKTOP-7QRHF82.5780.2 +3 -0
  22. aliceinwonderland/training_graph.json +3368 -0
  23. aliceinwonderland/training_graph.png +0 -0
  24. aliceinwonderland/training_log.json +19 -0
  25. aliceinwonderland/training_parameters.json +37 -0
  26. aliceinwonderland/training_prompt.json +3 -0
aliceinwonderland/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: models\Llama-2-13b-hf
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 0.8.2
aliceinwonderland/adapter_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "models\\Llama-2-13b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 64,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM",
26
+ "use_rslora": false
27
+ }
aliceinwonderland/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ec95cd40d9469ab7d56f444483c09b8727f0c834258363b46306bb4387fd5dd
3
+ size 104915722
aliceinwonderland/checkpoint-15-loss-1_17/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: models\Llama-2-13b-hf
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 0.8.2
aliceinwonderland/checkpoint-15-loss-1_17/adapter_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "models\\Llama-2-13b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 64,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM",
26
+ "use_rslora": false
27
+ }
aliceinwonderland/checkpoint-15-loss-1_17/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67856ecab2080beca8aca0a5e47b291f0805c418a022b4b2a97bb80d7f901ee7
3
+ size 104915722
aliceinwonderland/checkpoint-15-loss-1_17/training_log.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name": "Llama-2-13b-hf",
3
+ "base_model_class": "LlamaForCausalLM",
4
+ "base_loaded_in_4bit": true,
5
+ "base_loaded_in_8bit": false,
6
+ "projections": "q, v",
7
+ "loss": 1.1716,
8
+ "grad_norm": 1.0258234739303589,
9
+ "learning_rate": 1.3e-07,
10
+ "epoch": 0.13392857142857142,
11
+ "current_steps": 14,
12
+ "current_steps_adjusted": 14,
13
+ "epoch_adjusted": 0.13392857142857142,
14
+ "train_runtime": 60.8524,
15
+ "train_samples_per_second": 7.313,
16
+ "train_steps_per_second": 1.841,
17
+ "total_flos": 1819849670000640.0,
18
+ "train_loss": 0.7478187213773313
19
+ }
aliceinwonderland/checkpoint-15-loss-1_17/training_prompt.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "template_type": "raw_text"
3
+ }
aliceinwonderland/checkpoint-19-loss-0_90/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: models\Llama-2-13b-hf
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 0.8.2
aliceinwonderland/checkpoint-19-loss-0_90/adapter_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "models\\Llama-2-13b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 64,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM",
26
+ "use_rslora": false
27
+ }
aliceinwonderland/checkpoint-19-loss-0_90/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33728eded59d53993902b4320dd496b739df642eebfd71df1f2fded242f1cf5e
3
+ size 104915722
aliceinwonderland/checkpoint-19-loss-0_90/training_log.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name": "Llama-2-13b-hf",
3
+ "base_model_class": "LlamaForCausalLM",
4
+ "base_loaded_in_4bit": true,
5
+ "base_loaded_in_8bit": false,
6
+ "projections": "q, v",
7
+ "loss": 0.9004,
8
+ "grad_norm": 0.8880526423454285,
9
+ "learning_rate": 1.7000000000000001e-07,
10
+ "epoch": 0.16964285714285715,
11
+ "current_steps": 18,
12
+ "current_steps_adjusted": 18,
13
+ "epoch_adjusted": 0.16964285714285715,
14
+ "train_runtime": 60.8524,
15
+ "train_samples_per_second": 7.313,
16
+ "train_steps_per_second": 1.841,
17
+ "total_flos": 1819849670000640.0,
18
+ "train_loss": 0.7478187213773313
19
+ }
aliceinwonderland/checkpoint-19-loss-0_90/training_prompt.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "template_type": "raw_text"
3
+ }
aliceinwonderland/checkpoint-23-loss-0_60/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: models\Llama-2-13b-hf
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 0.8.2
aliceinwonderland/checkpoint-23-loss-0_60/adapter_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "models\\Llama-2-13b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 64,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM",
26
+ "use_rslora": false
27
+ }
aliceinwonderland/checkpoint-23-loss-0_60/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfb41f5ef71b08818902ba0f659dca42b6e428c3fca8dc04cccd89f096be730b
3
+ size 104915722
aliceinwonderland/checkpoint-23-loss-0_60/training_log.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name": "Llama-2-13b-hf",
3
+ "base_model_class": "LlamaForCausalLM",
4
+ "base_loaded_in_4bit": true,
5
+ "base_loaded_in_8bit": false,
6
+ "projections": "q, v",
7
+ "loss": 0.6049,
8
+ "grad_norm": 1.030413269996643,
9
+ "learning_rate": 2.0999999999999997e-07,
10
+ "epoch": 0.20535714285714285,
11
+ "current_steps": 22,
12
+ "current_steps_adjusted": 22,
13
+ "epoch_adjusted": 0.20535714285714285,
14
+ "train_runtime": 60.8524,
15
+ "train_samples_per_second": 7.313,
16
+ "train_steps_per_second": 1.841,
17
+ "total_flos": 1819849670000640.0,
18
+ "train_loss": 0.7478187213773313
19
+ }
aliceinwonderland/checkpoint-23-loss-0_60/training_prompt.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "template_type": "raw_text"
3
+ }
aliceinwonderland/runs/Jun04_00-27-53/events.out.tfevents.1717478875.DESKTOP-7QRHF82.5780.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d9c48c3ff261ea97428f58f99fdcc72538405b49e4ca84c9eb991bd5963f9b3
3
+ size 10741
aliceinwonderland/runs/Jun04_00-32-30/events.out.tfevents.1717479151.DESKTOP-7QRHF82.5780.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8acbbef9133cbacd0f4cc53e3cd9da98fe8510f03648ccc3a82c169bf8daa6c2
3
+ size 10326
aliceinwonderland/runs/Jun04_00-34-12/events.out.tfevents.1717479252.DESKTOP-7QRHF82.5780.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a194065673c018cd47b84d86a22c61ccebe7dc95cda5512cd1658d56cec6938
3
+ size 123223
aliceinwonderland/training_graph.json ADDED
@@ -0,0 +1,3368 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "current_steps": 0,
4
+ "loss": 0.6046,
5
+ "learning_rate": 1e-08,
6
+ "epoch": 0.008928571428571428
7
+ },
8
+ {
9
+ "current_steps": 1,
10
+ "loss": 0.6431,
11
+ "learning_rate": 2e-08,
12
+ "epoch": 0.017857142857142856
13
+ },
14
+ {
15
+ "current_steps": 2,
16
+ "loss": 0.6447,
17
+ "learning_rate": 3e-08,
18
+ "epoch": 0.026785714285714284
19
+ },
20
+ {
21
+ "current_steps": 3,
22
+ "loss": 0.7972,
23
+ "learning_rate": 4e-08,
24
+ "epoch": 0.03571428571428571
25
+ },
26
+ {
27
+ "current_steps": 4,
28
+ "loss": 0.6911,
29
+ "learning_rate": 5e-08,
30
+ "epoch": 0.044642857142857144
31
+ },
32
+ {
33
+ "current_steps": 5,
34
+ "loss": 0.8546,
35
+ "learning_rate": 6e-08,
36
+ "epoch": 0.05357142857142857
37
+ },
38
+ {
39
+ "current_steps": 6,
40
+ "loss": 0.7624,
41
+ "learning_rate": 6e-08,
42
+ "epoch": 0.0625
43
+ },
44
+ {
45
+ "current_steps": 7,
46
+ "loss": 0.6565,
47
+ "learning_rate": 7e-08,
48
+ "epoch": 0.07142857142857142
49
+ },
50
+ {
51
+ "current_steps": 8,
52
+ "loss": 0.6789,
53
+ "learning_rate": 8e-08,
54
+ "epoch": 0.08035714285714286
55
+ },
56
+ {
57
+ "current_steps": 9,
58
+ "loss": 0.8562,
59
+ "learning_rate": 8e-08,
60
+ "epoch": 0.08928571428571429
61
+ },
62
+ {
63
+ "current_steps": 10,
64
+ "loss": 0.8084,
65
+ "learning_rate": 9e-08,
66
+ "epoch": 0.09821428571428571
67
+ },
68
+ {
69
+ "current_steps": 11,
70
+ "loss": 0.7024,
71
+ "learning_rate": 1e-07,
72
+ "epoch": 0.10714285714285714
73
+ },
74
+ {
75
+ "current_steps": 12,
76
+ "loss": 0.7454,
77
+ "learning_rate": 1.0999999999999999e-07,
78
+ "epoch": 0.11607142857142858
79
+ },
80
+ {
81
+ "current_steps": 13,
82
+ "loss": 0.5896,
83
+ "learning_rate": 1.2e-07,
84
+ "epoch": 0.125
85
+ },
86
+ {
87
+ "current_steps": 14,
88
+ "loss": 1.1716,
89
+ "learning_rate": 1.3e-07,
90
+ "epoch": 0.13392857142857142
91
+ },
92
+ {
93
+ "current_steps": 15,
94
+ "loss": 0.8561,
95
+ "learning_rate": 1.4e-07,
96
+ "epoch": 0.14285714285714285
97
+ },
98
+ {
99
+ "current_steps": 16,
100
+ "loss": 0.9048,
101
+ "learning_rate": 1.5e-07,
102
+ "epoch": 0.15178571428571427
103
+ },
104
+ {
105
+ "current_steps": 17,
106
+ "loss": 0.6079,
107
+ "learning_rate": 1.6e-07,
108
+ "epoch": 0.16071428571428573
109
+ },
110
+ {
111
+ "current_steps": 18,
112
+ "loss": 0.9004,
113
+ "learning_rate": 1.7000000000000001e-07,
114
+ "epoch": 0.16964285714285715
115
+ },
116
+ {
117
+ "current_steps": 19,
118
+ "loss": 0.5512,
119
+ "learning_rate": 1.8e-07,
120
+ "epoch": 0.17857142857142858
121
+ },
122
+ {
123
+ "current_steps": 20,
124
+ "loss": 0.7782,
125
+ "learning_rate": 1.8999999999999998e-07,
126
+ "epoch": 0.1875
127
+ },
128
+ {
129
+ "current_steps": 21,
130
+ "loss": 0.7905,
131
+ "learning_rate": 2e-07,
132
+ "epoch": 0.19642857142857142
133
+ },
134
+ {
135
+ "current_steps": 22,
136
+ "loss": 0.6049,
137
+ "learning_rate": 2.0999999999999997e-07,
138
+ "epoch": 0.20535714285714285
139
+ },
140
+ {
141
+ "current_steps": 23,
142
+ "loss": 0.685,
143
+ "learning_rate": 2.1999999999999998e-07,
144
+ "epoch": 0.21428571428571427
145
+ },
146
+ {
147
+ "current_steps": 24,
148
+ "loss": 0.8171,
149
+ "learning_rate": 2.3e-07,
150
+ "epoch": 0.22321428571428573
151
+ },
152
+ {
153
+ "current_steps": 25,
154
+ "loss": 0.8018,
155
+ "learning_rate": 2.4e-07,
156
+ "epoch": 0.23214285714285715
157
+ },
158
+ {
159
+ "current_steps": 26,
160
+ "loss": 0.4959,
161
+ "learning_rate": 2.5e-07,
162
+ "epoch": 0.24107142857142858
163
+ },
164
+ {
165
+ "current_steps": 27,
166
+ "loss": 0.6348,
167
+ "learning_rate": 2.6e-07,
168
+ "epoch": 0.25
169
+ },
170
+ {
171
+ "current_steps": 28,
172
+ "loss": 0.8005,
173
+ "learning_rate": 2.7e-07,
174
+ "epoch": 0.25892857142857145
175
+ },
176
+ {
177
+ "current_steps": 29,
178
+ "loss": 0.6777,
179
+ "learning_rate": 2.8e-07,
180
+ "epoch": 0.26785714285714285
181
+ },
182
+ {
183
+ "current_steps": 30,
184
+ "loss": 0.9042,
185
+ "learning_rate": 2.9e-07,
186
+ "epoch": 0.2767857142857143
187
+ },
188
+ {
189
+ "current_steps": 31,
190
+ "loss": 0.6491,
191
+ "learning_rate": 3e-07,
192
+ "epoch": 0.2857142857142857
193
+ },
194
+ {
195
+ "current_steps": 32,
196
+ "loss": 1.0966,
197
+ "learning_rate": 3.1e-07,
198
+ "epoch": 0.29464285714285715
199
+ },
200
+ {
201
+ "current_steps": 33,
202
+ "loss": 0.7451,
203
+ "learning_rate": 3.2e-07,
204
+ "epoch": 0.30357142857142855
205
+ },
206
+ {
207
+ "current_steps": 34,
208
+ "loss": 1.1446,
209
+ "learning_rate": 3.2e-07,
210
+ "epoch": 0.3125
211
+ },
212
+ {
213
+ "current_steps": 35,
214
+ "loss": 0.7644,
215
+ "learning_rate": 3.3e-07,
216
+ "epoch": 0.32142857142857145
217
+ },
218
+ {
219
+ "current_steps": 36,
220
+ "loss": 0.7742,
221
+ "learning_rate": 3.4000000000000003e-07,
222
+ "epoch": 0.33035714285714285
223
+ },
224
+ {
225
+ "current_steps": 37,
226
+ "loss": 0.8247,
227
+ "learning_rate": 3.5e-07,
228
+ "epoch": 0.3392857142857143
229
+ },
230
+ {
231
+ "current_steps": 38,
232
+ "loss": 0.8667,
233
+ "learning_rate": 3.6e-07,
234
+ "epoch": 0.3482142857142857
235
+ },
236
+ {
237
+ "current_steps": 39,
238
+ "loss": 0.8309,
239
+ "learning_rate": 3.7e-07,
240
+ "epoch": 0.35714285714285715
241
+ },
242
+ {
243
+ "current_steps": 40,
244
+ "loss": 0.5913,
245
+ "learning_rate": 3.7999999999999996e-07,
246
+ "epoch": 0.36607142857142855
247
+ },
248
+ {
249
+ "current_steps": 41,
250
+ "loss": 0.5562,
251
+ "learning_rate": 3.8999999999999997e-07,
252
+ "epoch": 0.375
253
+ },
254
+ {
255
+ "current_steps": 42,
256
+ "loss": 1.6276,
257
+ "learning_rate": 4e-07,
258
+ "epoch": 0.38392857142857145
259
+ },
260
+ {
261
+ "current_steps": 43,
262
+ "loss": 0.682,
263
+ "learning_rate": 4.0999999999999994e-07,
264
+ "epoch": 0.39285714285714285
265
+ },
266
+ {
267
+ "current_steps": 44,
268
+ "loss": 0.8022,
269
+ "learning_rate": 4.1999999999999995e-07,
270
+ "epoch": 0.4017857142857143
271
+ },
272
+ {
273
+ "current_steps": 45,
274
+ "loss": 0.6702,
275
+ "learning_rate": 4.2999999999999996e-07,
276
+ "epoch": 0.4107142857142857
277
+ },
278
+ {
279
+ "current_steps": 46,
280
+ "loss": 0.6993,
281
+ "learning_rate": 4.3999999999999997e-07,
282
+ "epoch": 0.41964285714285715
283
+ },
284
+ {
285
+ "current_steps": 47,
286
+ "loss": 0.9685,
287
+ "learning_rate": 4.5e-07,
288
+ "epoch": 0.42857142857142855
289
+ },
290
+ {
291
+ "current_steps": 48,
292
+ "loss": 0.6637,
293
+ "learning_rate": 4.6e-07,
294
+ "epoch": 0.4375
295
+ },
296
+ {
297
+ "current_steps": 49,
298
+ "loss": 0.908,
299
+ "learning_rate": 4.6999999999999995e-07,
300
+ "epoch": 0.44642857142857145
301
+ },
302
+ {
303
+ "current_steps": 50,
304
+ "loss": 0.8683,
305
+ "learning_rate": 4.8e-07,
306
+ "epoch": 0.45535714285714285
307
+ },
308
+ {
309
+ "current_steps": 51,
310
+ "loss": 0.9243,
311
+ "learning_rate": 4.9e-07,
312
+ "epoch": 0.4642857142857143
313
+ },
314
+ {
315
+ "current_steps": 52,
316
+ "loss": 0.7933,
317
+ "learning_rate": 5e-07,
318
+ "epoch": 0.4732142857142857
319
+ },
320
+ {
321
+ "current_steps": 53,
322
+ "loss": 0.5856,
323
+ "learning_rate": 5.1e-07,
324
+ "epoch": 0.48214285714285715
325
+ },
326
+ {
327
+ "current_steps": 54,
328
+ "loss": 0.7097,
329
+ "learning_rate": 5.2e-07,
330
+ "epoch": 0.49107142857142855
331
+ },
332
+ {
333
+ "current_steps": 55,
334
+ "loss": 0.6476,
335
+ "learning_rate": 5.3e-07,
336
+ "epoch": 0.5
337
+ },
338
+ {
339
+ "current_steps": 56,
340
+ "loss": 0.8212,
341
+ "learning_rate": 5.4e-07,
342
+ "epoch": 0.5089285714285714
343
+ },
344
+ {
345
+ "current_steps": 57,
346
+ "loss": 0.7932,
347
+ "learning_rate": 5.5e-07,
348
+ "epoch": 0.5178571428571429
349
+ },
350
+ {
351
+ "current_steps": 58,
352
+ "loss": 0.8155,
353
+ "learning_rate": 5.6e-07,
354
+ "epoch": 0.5267857142857143
355
+ },
356
+ {
357
+ "current_steps": 59,
358
+ "loss": 0.5644,
359
+ "learning_rate": 5.699999999999999e-07,
360
+ "epoch": 0.5357142857142857
361
+ },
362
+ {
363
+ "current_steps": 60,
364
+ "loss": 0.8935,
365
+ "learning_rate": 5.8e-07,
366
+ "epoch": 0.5446428571428571
367
+ },
368
+ {
369
+ "current_steps": 61,
370
+ "loss": 0.6935,
371
+ "learning_rate": 5.9e-07,
372
+ "epoch": 0.5535714285714286
373
+ },
374
+ {
375
+ "current_steps": 62,
376
+ "loss": 0.6186,
377
+ "learning_rate": 6e-07,
378
+ "epoch": 0.5625
379
+ },
380
+ {
381
+ "current_steps": 63,
382
+ "loss": 0.7528,
383
+ "learning_rate": 6.1e-07,
384
+ "epoch": 0.5714285714285714
385
+ },
386
+ {
387
+ "current_steps": 64,
388
+ "loss": 0.7043,
389
+ "learning_rate": 6.2e-07,
390
+ "epoch": 0.5803571428571429
391
+ },
392
+ {
393
+ "current_steps": 65,
394
+ "loss": 0.5926,
395
+ "learning_rate": 6.3e-07,
396
+ "epoch": 0.5892857142857143
397
+ },
398
+ {
399
+ "current_steps": 66,
400
+ "loss": 0.7927,
401
+ "learning_rate": 6.4e-07,
402
+ "epoch": 0.5982142857142857
403
+ },
404
+ {
405
+ "current_steps": 67,
406
+ "loss": 0.5625,
407
+ "learning_rate": 6.5e-07,
408
+ "epoch": 0.6071428571428571
409
+ },
410
+ {
411
+ "current_steps": 68,
412
+ "loss": 0.707,
413
+ "learning_rate": 6.6e-07,
414
+ "epoch": 0.6160714285714286
415
+ },
416
+ {
417
+ "current_steps": 69,
418
+ "loss": 0.7023,
419
+ "learning_rate": 6.7e-07,
420
+ "epoch": 0.625
421
+ },
422
+ {
423
+ "current_steps": 70,
424
+ "loss": 0.586,
425
+ "learning_rate": 6.800000000000001e-07,
426
+ "epoch": 0.6339285714285714
427
+ },
428
+ {
429
+ "current_steps": 71,
430
+ "loss": 0.5741,
431
+ "learning_rate": 6.9e-07,
432
+ "epoch": 0.6428571428571429
433
+ },
434
+ {
435
+ "current_steps": 72,
436
+ "loss": 1.086,
437
+ "learning_rate": 7e-07,
438
+ "epoch": 0.6517857142857143
439
+ },
440
+ {
441
+ "current_steps": 73,
442
+ "loss": 0.6381,
443
+ "learning_rate": 7.1e-07,
444
+ "epoch": 0.6607142857142857
445
+ },
446
+ {
447
+ "current_steps": 74,
448
+ "loss": 0.7509,
449
+ "learning_rate": 7.2e-07,
450
+ "epoch": 0.6696428571428571
451
+ },
452
+ {
453
+ "current_steps": 75,
454
+ "loss": 0.8276,
455
+ "learning_rate": 7.3e-07,
456
+ "epoch": 0.6785714285714286
457
+ },
458
+ {
459
+ "current_steps": 76,
460
+ "loss": 0.7623,
461
+ "learning_rate": 7.4e-07,
462
+ "epoch": 0.6875
463
+ },
464
+ {
465
+ "current_steps": 77,
466
+ "loss": 0.9499,
467
+ "learning_rate": 7.5e-07,
468
+ "epoch": 0.6964285714285714
469
+ },
470
+ {
471
+ "current_steps": 78,
472
+ "loss": 0.8563,
473
+ "learning_rate": 7.599999999999999e-07,
474
+ "epoch": 0.7053571428571429
475
+ },
476
+ {
477
+ "current_steps": 79,
478
+ "loss": 0.6512,
479
+ "learning_rate": 7.699999999999999e-07,
480
+ "epoch": 0.7142857142857143
481
+ },
482
+ {
483
+ "current_steps": 80,
484
+ "loss": 0.843,
485
+ "learning_rate": 7.799999999999999e-07,
486
+ "epoch": 0.7232142857142857
487
+ },
488
+ {
489
+ "current_steps": 81,
490
+ "loss": 0.7272,
491
+ "learning_rate": 7.9e-07,
492
+ "epoch": 0.7321428571428571
493
+ },
494
+ {
495
+ "current_steps": 82,
496
+ "loss": 0.5161,
497
+ "learning_rate": 8e-07,
498
+ "epoch": 0.7410714285714286
499
+ },
500
+ {
501
+ "current_steps": 83,
502
+ "loss": 0.8293,
503
+ "learning_rate": 8.1e-07,
504
+ "epoch": 0.75
505
+ },
506
+ {
507
+ "current_steps": 84,
508
+ "loss": 0.8704,
509
+ "learning_rate": 8.199999999999999e-07,
510
+ "epoch": 0.7589285714285714
511
+ },
512
+ {
513
+ "current_steps": 85,
514
+ "loss": 0.7255,
515
+ "learning_rate": 8.299999999999999e-07,
516
+ "epoch": 0.7678571428571429
517
+ },
518
+ {
519
+ "current_steps": 86,
520
+ "loss": 0.6252,
521
+ "learning_rate": 8.399999999999999e-07,
522
+ "epoch": 0.7767857142857143
523
+ },
524
+ {
525
+ "current_steps": 87,
526
+ "loss": 0.8116,
527
+ "learning_rate": 8.499999999999999e-07,
528
+ "epoch": 0.7857142857142857
529
+ },
530
+ {
531
+ "current_steps": 88,
532
+ "loss": 0.7703,
533
+ "learning_rate": 8.599999999999999e-07,
534
+ "epoch": 0.7946428571428571
535
+ },
536
+ {
537
+ "current_steps": 89,
538
+ "loss": 0.6496,
539
+ "learning_rate": 8.699999999999999e-07,
540
+ "epoch": 0.8035714285714286
541
+ },
542
+ {
543
+ "current_steps": 90,
544
+ "loss": 0.8585,
545
+ "learning_rate": 8.799999999999999e-07,
546
+ "epoch": 0.8125
547
+ },
548
+ {
549
+ "current_steps": 91,
550
+ "loss": 0.905,
551
+ "learning_rate": 8.9e-07,
552
+ "epoch": 0.8214285714285714
553
+ },
554
+ {
555
+ "current_steps": 92,
556
+ "loss": 0.9139,
557
+ "learning_rate": 9e-07,
558
+ "epoch": 0.8303571428571429
559
+ },
560
+ {
561
+ "current_steps": 93,
562
+ "loss": 0.9925,
563
+ "learning_rate": 9.1e-07,
564
+ "epoch": 0.8392857142857143
565
+ },
566
+ {
567
+ "current_steps": 94,
568
+ "loss": 0.7344,
569
+ "learning_rate": 9.2e-07,
570
+ "epoch": 0.8482142857142857
571
+ },
572
+ {
573
+ "current_steps": 95,
574
+ "loss": 0.7477,
575
+ "learning_rate": 9.3e-07,
576
+ "epoch": 0.8571428571428571
577
+ },
578
+ {
579
+ "current_steps": 96,
580
+ "loss": 0.671,
581
+ "learning_rate": 9.399999999999999e-07,
582
+ "epoch": 0.8660714285714286
583
+ },
584
+ {
585
+ "current_steps": 97,
586
+ "loss": 0.9654,
587
+ "learning_rate": 9.499999999999999e-07,
588
+ "epoch": 0.875
589
+ },
590
+ {
591
+ "current_steps": 98,
592
+ "loss": 0.6788,
593
+ "learning_rate": 9.6e-07,
594
+ "epoch": 0.8839285714285714
595
+ },
596
+ {
597
+ "current_steps": 99,
598
+ "loss": 0.764,
599
+ "learning_rate": 9.7e-07,
600
+ "epoch": 0.8928571428571429
601
+ },
602
+ {
603
+ "current_steps": 100,
604
+ "loss": 0.7536,
605
+ "learning_rate": 9.8e-07,
606
+ "epoch": 0.9017857142857143
607
+ },
608
+ {
609
+ "current_steps": 101,
610
+ "loss": 0.6409,
611
+ "learning_rate": 9.9e-07,
612
+ "epoch": 0.9107142857142857
613
+ },
614
+ {
615
+ "current_steps": 102,
616
+ "loss": 0.904,
617
+ "learning_rate": 1e-06,
618
+ "epoch": 0.9196428571428571
619
+ },
620
+ {
621
+ "current_steps": 103,
622
+ "loss": 0.7079,
623
+ "learning_rate": 9.978260869565217e-07,
624
+ "epoch": 0.9285714285714286
625
+ },
626
+ {
627
+ "current_steps": 104,
628
+ "loss": 0.748,
629
+ "learning_rate": 9.956521739130434e-07,
630
+ "epoch": 0.9375
631
+ },
632
+ {
633
+ "current_steps": 105,
634
+ "loss": 0.7228,
635
+ "learning_rate": 9.934782608695653e-07,
636
+ "epoch": 0.9464285714285714
637
+ },
638
+ {
639
+ "current_steps": 106,
640
+ "loss": 0.722,
641
+ "learning_rate": 9.91304347826087e-07,
642
+ "epoch": 0.9553571428571429
643
+ },
644
+ {
645
+ "current_steps": 107,
646
+ "loss": 0.8011,
647
+ "learning_rate": 9.891304347826085e-07,
648
+ "epoch": 0.9642857142857143
649
+ },
650
+ {
651
+ "current_steps": 108,
652
+ "loss": 0.8125,
653
+ "learning_rate": 9.869565217391304e-07,
654
+ "epoch": 0.9732142857142857
655
+ },
656
+ {
657
+ "current_steps": 109,
658
+ "loss": 0.8091,
659
+ "learning_rate": 9.847826086956522e-07,
660
+ "epoch": 0.9821428571428571
661
+ },
662
+ {
663
+ "current_steps": 110,
664
+ "loss": 0.9399,
665
+ "learning_rate": 9.826086956521739e-07,
666
+ "epoch": 0.9910714285714286
667
+ },
668
+ {
669
+ "current_steps": 111,
670
+ "loss": 1.0917,
671
+ "learning_rate": 9.804347826086956e-07,
672
+ "epoch": 1.0
673
+ },
674
+ {
675
+ "current_steps": 112,
676
+ "loss": 0.9014,
677
+ "learning_rate": 9.782608695652173e-07,
678
+ "epoch": 1.0089285714285714
679
+ },
680
+ {
681
+ "current_steps": 113,
682
+ "loss": 0.873,
683
+ "learning_rate": 9.782608695652173e-07,
684
+ "epoch": 1.0178571428571428
685
+ },
686
+ {
687
+ "current_steps": 114,
688
+ "loss": 0.7153,
689
+ "learning_rate": 9.76086956521739e-07,
690
+ "epoch": 1.0267857142857142
691
+ },
692
+ {
693
+ "current_steps": 115,
694
+ "loss": 0.8828,
695
+ "learning_rate": 9.73913043478261e-07,
696
+ "epoch": 1.0357142857142858
697
+ },
698
+ {
699
+ "current_steps": 116,
700
+ "loss": 1.0329,
701
+ "learning_rate": 9.717391304347827e-07,
702
+ "epoch": 1.0446428571428572
703
+ },
704
+ {
705
+ "current_steps": 117,
706
+ "loss": 1.057,
707
+ "learning_rate": 9.695652173913042e-07,
708
+ "epoch": 1.0535714285714286
709
+ },
710
+ {
711
+ "current_steps": 118,
712
+ "loss": 0.8047,
713
+ "learning_rate": 9.67391304347826e-07,
714
+ "epoch": 1.0625
715
+ },
716
+ {
717
+ "current_steps": 119,
718
+ "loss": 0.7098,
719
+ "learning_rate": 9.652173913043478e-07,
720
+ "epoch": 1.0714285714285714
721
+ },
722
+ {
723
+ "current_steps": 120,
724
+ "loss": 1.094,
725
+ "learning_rate": 9.630434782608695e-07,
726
+ "epoch": 1.0803571428571428
727
+ },
728
+ {
729
+ "current_steps": 121,
730
+ "loss": 0.7521,
731
+ "learning_rate": 9.608695652173912e-07,
732
+ "epoch": 1.0892857142857142
733
+ },
734
+ {
735
+ "current_steps": 122,
736
+ "loss": 0.9738,
737
+ "learning_rate": 9.58695652173913e-07,
738
+ "epoch": 1.0982142857142858
739
+ },
740
+ {
741
+ "current_steps": 123,
742
+ "loss": 0.5577,
743
+ "learning_rate": 9.565217391304349e-07,
744
+ "epoch": 1.1071428571428572
745
+ },
746
+ {
747
+ "current_steps": 124,
748
+ "loss": 1.046,
749
+ "learning_rate": 9.543478260869566e-07,
750
+ "epoch": 1.1160714285714286
751
+ },
752
+ {
753
+ "current_steps": 125,
754
+ "loss": 0.597,
755
+ "learning_rate": 9.521739130434783e-07,
756
+ "epoch": 1.125
757
+ },
758
+ {
759
+ "current_steps": 126,
760
+ "loss": 0.7996,
761
+ "learning_rate": 9.499999999999999e-07,
762
+ "epoch": 1.1339285714285714
763
+ },
764
+ {
765
+ "current_steps": 127,
766
+ "loss": 0.9885,
767
+ "learning_rate": 9.478260869565216e-07,
768
+ "epoch": 1.1428571428571428
769
+ },
770
+ {
771
+ "current_steps": 128,
772
+ "loss": 0.6274,
773
+ "learning_rate": 9.456521739130434e-07,
774
+ "epoch": 1.1517857142857142
775
+ },
776
+ {
777
+ "current_steps": 129,
778
+ "loss": 0.8557,
779
+ "learning_rate": 9.434782608695652e-07,
780
+ "epoch": 1.1607142857142858
781
+ },
782
+ {
783
+ "current_steps": 130,
784
+ "loss": 0.702,
785
+ "learning_rate": 9.41304347826087e-07,
786
+ "epoch": 1.1696428571428572
787
+ },
788
+ {
789
+ "current_steps": 131,
790
+ "loss": 0.6905,
791
+ "learning_rate": 9.391304347826087e-07,
792
+ "epoch": 1.1785714285714286
793
+ },
794
+ {
795
+ "current_steps": 132,
796
+ "loss": 0.5707,
797
+ "learning_rate": 9.369565217391304e-07,
798
+ "epoch": 1.1875
799
+ },
800
+ {
801
+ "current_steps": 133,
802
+ "loss": 0.6121,
803
+ "learning_rate": 9.347826086956522e-07,
804
+ "epoch": 1.1964285714285714
805
+ },
806
+ {
807
+ "current_steps": 134,
808
+ "loss": 0.8348,
809
+ "learning_rate": 9.326086956521738e-07,
810
+ "epoch": 1.2053571428571428
811
+ },
812
+ {
813
+ "current_steps": 135,
814
+ "loss": 0.8768,
815
+ "learning_rate": 9.304347826086955e-07,
816
+ "epoch": 1.2142857142857142
817
+ },
818
+ {
819
+ "current_steps": 136,
820
+ "loss": 0.5648,
821
+ "learning_rate": 9.282608695652174e-07,
822
+ "epoch": 1.2232142857142858
823
+ },
824
+ {
825
+ "current_steps": 137,
826
+ "loss": 0.6316,
827
+ "learning_rate": 9.260869565217391e-07,
828
+ "epoch": 1.2321428571428572
829
+ },
830
+ {
831
+ "current_steps": 138,
832
+ "loss": 1.1728,
833
+ "learning_rate": 9.239130434782608e-07,
834
+ "epoch": 1.2410714285714286
835
+ },
836
+ {
837
+ "current_steps": 139,
838
+ "loss": 0.7299,
839
+ "learning_rate": 9.217391304347826e-07,
840
+ "epoch": 1.25
841
+ },
842
+ {
843
+ "current_steps": 140,
844
+ "loss": 0.6284,
845
+ "learning_rate": 9.195652173913043e-07,
846
+ "epoch": 1.2589285714285714
847
+ },
848
+ {
849
+ "current_steps": 141,
850
+ "loss": 0.6366,
851
+ "learning_rate": 9.17391304347826e-07,
852
+ "epoch": 1.2678571428571428
853
+ },
854
+ {
855
+ "current_steps": 142,
856
+ "loss": 0.7357,
857
+ "learning_rate": 9.152173913043479e-07,
858
+ "epoch": 1.2767857142857144
859
+ },
860
+ {
861
+ "current_steps": 143,
862
+ "loss": 0.8618,
863
+ "learning_rate": 9.130434782608695e-07,
864
+ "epoch": 1.2857142857142856
865
+ },
866
+ {
867
+ "current_steps": 144,
868
+ "loss": 0.6803,
869
+ "learning_rate": 9.108695652173912e-07,
870
+ "epoch": 1.2946428571428572
871
+ },
872
+ {
873
+ "current_steps": 145,
874
+ "loss": 0.8093,
875
+ "learning_rate": 9.08695652173913e-07,
876
+ "epoch": 1.3035714285714286
877
+ },
878
+ {
879
+ "current_steps": 146,
880
+ "loss": 0.6808,
881
+ "learning_rate": 9.065217391304347e-07,
882
+ "epoch": 1.3125
883
+ },
884
+ {
885
+ "current_steps": 147,
886
+ "loss": 0.7173,
887
+ "learning_rate": 9.043478260869564e-07,
888
+ "epoch": 1.3214285714285714
889
+ },
890
+ {
891
+ "current_steps": 148,
892
+ "loss": 0.6964,
893
+ "learning_rate": 9.021739130434782e-07,
894
+ "epoch": 1.3303571428571428
895
+ },
896
+ {
897
+ "current_steps": 149,
898
+ "loss": 0.5458,
899
+ "learning_rate": 9e-07,
900
+ "epoch": 1.3392857142857144
901
+ },
902
+ {
903
+ "current_steps": 150,
904
+ "loss": 0.5362,
905
+ "learning_rate": 8.978260869565218e-07,
906
+ "epoch": 1.3482142857142856
907
+ },
908
+ {
909
+ "current_steps": 151,
910
+ "loss": 0.7248,
911
+ "learning_rate": 8.956521739130435e-07,
912
+ "epoch": 1.3571428571428572
913
+ },
914
+ {
915
+ "current_steps": 152,
916
+ "loss": 0.9701,
917
+ "learning_rate": 8.934782608695651e-07,
918
+ "epoch": 1.3660714285714286
919
+ },
920
+ {
921
+ "current_steps": 153,
922
+ "loss": 0.6072,
923
+ "learning_rate": 8.913043478260869e-07,
924
+ "epoch": 1.375
925
+ },
926
+ {
927
+ "current_steps": 154,
928
+ "loss": 0.8135,
929
+ "learning_rate": 8.891304347826086e-07,
930
+ "epoch": 1.3839285714285714
931
+ },
932
+ {
933
+ "current_steps": 155,
934
+ "loss": 0.6519,
935
+ "learning_rate": 8.869565217391303e-07,
936
+ "epoch": 1.3928571428571428
937
+ },
938
+ {
939
+ "current_steps": 156,
940
+ "loss": 0.7911,
941
+ "learning_rate": 8.847826086956522e-07,
942
+ "epoch": 1.4017857142857144
943
+ },
944
+ {
945
+ "current_steps": 157,
946
+ "loss": 0.7084,
947
+ "learning_rate": 8.826086956521739e-07,
948
+ "epoch": 1.4107142857142856
949
+ },
950
+ {
951
+ "current_steps": 158,
952
+ "loss": 0.6062,
953
+ "learning_rate": 8.804347826086956e-07,
954
+ "epoch": 1.4196428571428572
955
+ },
956
+ {
957
+ "current_steps": 159,
958
+ "loss": 0.5372,
959
+ "learning_rate": 8.782608695652174e-07,
960
+ "epoch": 1.4285714285714286
961
+ },
962
+ {
963
+ "current_steps": 160,
964
+ "loss": 0.7001,
965
+ "learning_rate": 8.760869565217391e-07,
966
+ "epoch": 1.4375
967
+ },
968
+ {
969
+ "current_steps": 161,
970
+ "loss": 0.628,
971
+ "learning_rate": 8.739130434782607e-07,
972
+ "epoch": 1.4464285714285714
973
+ },
974
+ {
975
+ "current_steps": 162,
976
+ "loss": 0.6766,
977
+ "learning_rate": 8.717391304347826e-07,
978
+ "epoch": 1.4553571428571428
979
+ },
980
+ {
981
+ "current_steps": 163,
982
+ "loss": 0.7406,
983
+ "learning_rate": 8.695652173913043e-07,
984
+ "epoch": 1.4642857142857144
985
+ },
986
+ {
987
+ "current_steps": 164,
988
+ "loss": 0.7032,
989
+ "learning_rate": 8.67391304347826e-07,
990
+ "epoch": 1.4732142857142856
991
+ },
992
+ {
993
+ "current_steps": 165,
994
+ "loss": 0.8338,
995
+ "learning_rate": 8.652173913043478e-07,
996
+ "epoch": 1.4821428571428572
997
+ },
998
+ {
999
+ "current_steps": 166,
1000
+ "loss": 0.6067,
1001
+ "learning_rate": 8.630434782608695e-07,
1002
+ "epoch": 1.4910714285714286
1003
+ },
1004
+ {
1005
+ "current_steps": 167,
1006
+ "loss": 0.6988,
1007
+ "learning_rate": 8.608695652173913e-07,
1008
+ "epoch": 1.5
1009
+ },
1010
+ {
1011
+ "current_steps": 168,
1012
+ "loss": 0.6294,
1013
+ "learning_rate": 8.586956521739131e-07,
1014
+ "epoch": 1.5089285714285714
1015
+ },
1016
+ {
1017
+ "current_steps": 169,
1018
+ "loss": 0.7358,
1019
+ "learning_rate": 8.565217391304348e-07,
1020
+ "epoch": 1.5178571428571428
1021
+ },
1022
+ {
1023
+ "current_steps": 170,
1024
+ "loss": 0.7709,
1025
+ "learning_rate": 8.543478260869565e-07,
1026
+ "epoch": 1.5267857142857144
1027
+ },
1028
+ {
1029
+ "current_steps": 171,
1030
+ "loss": 0.8913,
1031
+ "learning_rate": 8.521739130434782e-07,
1032
+ "epoch": 1.5357142857142856
1033
+ },
1034
+ {
1035
+ "current_steps": 172,
1036
+ "loss": 0.697,
1037
+ "learning_rate": 8.499999999999999e-07,
1038
+ "epoch": 1.5446428571428572
1039
+ },
1040
+ {
1041
+ "current_steps": 173,
1042
+ "loss": 0.7902,
1043
+ "learning_rate": 8.478260869565217e-07,
1044
+ "epoch": 1.5535714285714286
1045
+ },
1046
+ {
1047
+ "current_steps": 174,
1048
+ "loss": 0.7858,
1049
+ "learning_rate": 8.456521739130434e-07,
1050
+ "epoch": 1.5625
1051
+ },
1052
+ {
1053
+ "current_steps": 175,
1054
+ "loss": 0.8903,
1055
+ "learning_rate": 8.434782608695652e-07,
1056
+ "epoch": 1.5714285714285714
1057
+ },
1058
+ {
1059
+ "current_steps": 176,
1060
+ "loss": 0.8324,
1061
+ "learning_rate": 8.41304347826087e-07,
1062
+ "epoch": 1.5803571428571428
1063
+ },
1064
+ {
1065
+ "current_steps": 177,
1066
+ "loss": 0.7323,
1067
+ "learning_rate": 8.391304347826087e-07,
1068
+ "epoch": 1.5892857142857144
1069
+ },
1070
+ {
1071
+ "current_steps": 178,
1072
+ "loss": 0.7527,
1073
+ "learning_rate": 8.369565217391304e-07,
1074
+ "epoch": 1.5982142857142856
1075
+ },
1076
+ {
1077
+ "current_steps": 179,
1078
+ "loss": 0.8336,
1079
+ "learning_rate": 8.347826086956521e-07,
1080
+ "epoch": 1.6071428571428572
1081
+ },
1082
+ {
1083
+ "current_steps": 180,
1084
+ "loss": 0.7886,
1085
+ "learning_rate": 8.326086956521738e-07,
1086
+ "epoch": 1.6160714285714286
1087
+ },
1088
+ {
1089
+ "current_steps": 181,
1090
+ "loss": 0.7455,
1091
+ "learning_rate": 8.304347826086955e-07,
1092
+ "epoch": 1.625
1093
+ },
1094
+ {
1095
+ "current_steps": 182,
1096
+ "loss": 0.7702,
1097
+ "learning_rate": 8.282608695652174e-07,
1098
+ "epoch": 1.6339285714285714
1099
+ },
1100
+ {
1101
+ "current_steps": 183,
1102
+ "loss": 0.6935,
1103
+ "learning_rate": 8.260869565217391e-07,
1104
+ "epoch": 1.6428571428571428
1105
+ },
1106
+ {
1107
+ "current_steps": 184,
1108
+ "loss": 0.6778,
1109
+ "learning_rate": 8.239130434782609e-07,
1110
+ "epoch": 1.6517857142857144
1111
+ },
1112
+ {
1113
+ "current_steps": 185,
1114
+ "loss": 0.7623,
1115
+ "learning_rate": 8.217391304347826e-07,
1116
+ "epoch": 1.6607142857142856
1117
+ },
1118
+ {
1119
+ "current_steps": 186,
1120
+ "loss": 0.8068,
1121
+ "learning_rate": 8.195652173913043e-07,
1122
+ "epoch": 1.6696428571428572
1123
+ },
1124
+ {
1125
+ "current_steps": 187,
1126
+ "loss": 0.6384,
1127
+ "learning_rate": 8.173913043478261e-07,
1128
+ "epoch": 1.6785714285714286
1129
+ },
1130
+ {
1131
+ "current_steps": 188,
1132
+ "loss": 0.9876,
1133
+ "learning_rate": 8.152173913043478e-07,
1134
+ "epoch": 1.6875
1135
+ },
1136
+ {
1137
+ "current_steps": 189,
1138
+ "loss": 0.5316,
1139
+ "learning_rate": 8.130434782608695e-07,
1140
+ "epoch": 1.6964285714285714
1141
+ },
1142
+ {
1143
+ "current_steps": 190,
1144
+ "loss": 0.6117,
1145
+ "learning_rate": 8.108695652173913e-07,
1146
+ "epoch": 1.7053571428571428
1147
+ },
1148
+ {
1149
+ "current_steps": 191,
1150
+ "loss": 0.5897,
1151
+ "learning_rate": 8.08695652173913e-07,
1152
+ "epoch": 1.7142857142857144
1153
+ },
1154
+ {
1155
+ "current_steps": 192,
1156
+ "loss": 0.7045,
1157
+ "learning_rate": 8.065217391304347e-07,
1158
+ "epoch": 1.7232142857142856
1159
+ },
1160
+ {
1161
+ "current_steps": 193,
1162
+ "loss": 0.7491,
1163
+ "learning_rate": 8.043478260869565e-07,
1164
+ "epoch": 1.7321428571428572
1165
+ },
1166
+ {
1167
+ "current_steps": 194,
1168
+ "loss": 0.8067,
1169
+ "learning_rate": 8.021739130434782e-07,
1170
+ "epoch": 1.7410714285714286
1171
+ },
1172
+ {
1173
+ "current_steps": 195,
1174
+ "loss": 0.9085,
1175
+ "learning_rate": 8e-07,
1176
+ "epoch": 1.75
1177
+ },
1178
+ {
1179
+ "current_steps": 196,
1180
+ "loss": 0.7977,
1181
+ "learning_rate": 7.978260869565217e-07,
1182
+ "epoch": 1.7589285714285714
1183
+ },
1184
+ {
1185
+ "current_steps": 197,
1186
+ "loss": 0.7509,
1187
+ "learning_rate": 7.956521739130434e-07,
1188
+ "epoch": 1.7678571428571428
1189
+ },
1190
+ {
1191
+ "current_steps": 198,
1192
+ "loss": 0.7048,
1193
+ "learning_rate": 7.934782608695651e-07,
1194
+ "epoch": 1.7767857142857144
1195
+ },
1196
+ {
1197
+ "current_steps": 199,
1198
+ "loss": 0.6452,
1199
+ "learning_rate": 7.913043478260869e-07,
1200
+ "epoch": 1.7857142857142856
1201
+ },
1202
+ {
1203
+ "current_steps": 200,
1204
+ "loss": 0.7265,
1205
+ "learning_rate": 7.891304347826086e-07,
1206
+ "epoch": 1.7946428571428572
1207
+ },
1208
+ {
1209
+ "current_steps": 201,
1210
+ "loss": 0.7936,
1211
+ "learning_rate": 7.869565217391305e-07,
1212
+ "epoch": 1.8035714285714286
1213
+ },
1214
+ {
1215
+ "current_steps": 202,
1216
+ "loss": 0.7336,
1217
+ "learning_rate": 7.847826086956522e-07,
1218
+ "epoch": 1.8125
1219
+ },
1220
+ {
1221
+ "current_steps": 203,
1222
+ "loss": 0.6462,
1223
+ "learning_rate": 7.826086956521739e-07,
1224
+ "epoch": 1.8214285714285714
1225
+ },
1226
+ {
1227
+ "current_steps": 204,
1228
+ "loss": 0.579,
1229
+ "learning_rate": 7.804347826086957e-07,
1230
+ "epoch": 1.8303571428571428
1231
+ },
1232
+ {
1233
+ "current_steps": 205,
1234
+ "loss": 0.6014,
1235
+ "learning_rate": 7.782608695652173e-07,
1236
+ "epoch": 1.8392857142857144
1237
+ },
1238
+ {
1239
+ "current_steps": 206,
1240
+ "loss": 0.684,
1241
+ "learning_rate": 7.76086956521739e-07,
1242
+ "epoch": 1.8482142857142856
1243
+ },
1244
+ {
1245
+ "current_steps": 207,
1246
+ "loss": 0.5932,
1247
+ "learning_rate": 7.739130434782608e-07,
1248
+ "epoch": 1.8571428571428572
1249
+ },
1250
+ {
1251
+ "current_steps": 208,
1252
+ "loss": 0.7736,
1253
+ "learning_rate": 7.717391304347826e-07,
1254
+ "epoch": 1.8660714285714286
1255
+ },
1256
+ {
1257
+ "current_steps": 209,
1258
+ "loss": 0.7601,
1259
+ "learning_rate": 7.695652173913043e-07,
1260
+ "epoch": 1.875
1261
+ },
1262
+ {
1263
+ "current_steps": 210,
1264
+ "loss": 0.8428,
1265
+ "learning_rate": 7.673913043478261e-07,
1266
+ "epoch": 1.8839285714285714
1267
+ },
1268
+ {
1269
+ "current_steps": 211,
1270
+ "loss": 0.8017,
1271
+ "learning_rate": 7.652173913043478e-07,
1272
+ "epoch": 1.8928571428571428
1273
+ },
1274
+ {
1275
+ "current_steps": 212,
1276
+ "loss": 0.5998,
1277
+ "learning_rate": 7.630434782608695e-07,
1278
+ "epoch": 1.9017857142857144
1279
+ },
1280
+ {
1281
+ "current_steps": 213,
1282
+ "loss": 0.9071,
1283
+ "learning_rate": 7.608695652173913e-07,
1284
+ "epoch": 1.9107142857142856
1285
+ },
1286
+ {
1287
+ "current_steps": 214,
1288
+ "loss": 0.8255,
1289
+ "learning_rate": 7.58695652173913e-07,
1290
+ "epoch": 1.9196428571428572
1291
+ },
1292
+ {
1293
+ "current_steps": 215,
1294
+ "loss": 0.9256,
1295
+ "learning_rate": 7.565217391304347e-07,
1296
+ "epoch": 1.9285714285714286
1297
+ },
1298
+ {
1299
+ "current_steps": 216,
1300
+ "loss": 0.6745,
1301
+ "learning_rate": 7.543478260869565e-07,
1302
+ "epoch": 1.9375
1303
+ },
1304
+ {
1305
+ "current_steps": 217,
1306
+ "loss": 0.6372,
1307
+ "learning_rate": 7.521739130434782e-07,
1308
+ "epoch": 1.9464285714285714
1309
+ },
1310
+ {
1311
+ "current_steps": 218,
1312
+ "loss": 0.6495,
1313
+ "learning_rate": 7.5e-07,
1314
+ "epoch": 1.9553571428571428
1315
+ },
1316
+ {
1317
+ "current_steps": 219,
1318
+ "loss": 0.6054,
1319
+ "learning_rate": 7.478260869565217e-07,
1320
+ "epoch": 1.9642857142857144
1321
+ },
1322
+ {
1323
+ "current_steps": 220,
1324
+ "loss": 0.9751,
1325
+ "learning_rate": 7.478260869565217e-07,
1326
+ "epoch": 1.9732142857142856
1327
+ },
1328
+ {
1329
+ "current_steps": 221,
1330
+ "loss": 0.6258,
1331
+ "learning_rate": 7.456521739130434e-07,
1332
+ "epoch": 1.9821428571428572
1333
+ },
1334
+ {
1335
+ "current_steps": 222,
1336
+ "loss": 0.794,
1337
+ "learning_rate": 7.434782608695653e-07,
1338
+ "epoch": 1.9910714285714286
1339
+ },
1340
+ {
1341
+ "current_steps": 223,
1342
+ "loss": 0.9991,
1343
+ "learning_rate": 7.41304347826087e-07,
1344
+ "epoch": 2.0
1345
+ },
1346
+ {
1347
+ "current_steps": 224,
1348
+ "loss": 0.8048,
1349
+ "learning_rate": 7.391304347826086e-07,
1350
+ "epoch": 2.0089285714285716
1351
+ },
1352
+ {
1353
+ "current_steps": 225,
1354
+ "loss": 0.8439,
1355
+ "learning_rate": 7.369565217391304e-07,
1356
+ "epoch": 2.017857142857143
1357
+ },
1358
+ {
1359
+ "current_steps": 226,
1360
+ "loss": 0.7546,
1361
+ "learning_rate": 7.347826086956521e-07,
1362
+ "epoch": 2.0267857142857144
1363
+ },
1364
+ {
1365
+ "current_steps": 227,
1366
+ "loss": 0.8195,
1367
+ "learning_rate": 7.326086956521738e-07,
1368
+ "epoch": 2.0357142857142856
1369
+ },
1370
+ {
1371
+ "current_steps": 228,
1372
+ "loss": 0.6988,
1373
+ "learning_rate": 7.304347826086957e-07,
1374
+ "epoch": 2.044642857142857
1375
+ },
1376
+ {
1377
+ "current_steps": 229,
1378
+ "loss": 0.8419,
1379
+ "learning_rate": 7.282608695652174e-07,
1380
+ "epoch": 2.0535714285714284
1381
+ },
1382
+ {
1383
+ "current_steps": 230,
1384
+ "loss": 0.6133,
1385
+ "learning_rate": 7.260869565217391e-07,
1386
+ "epoch": 2.0625
1387
+ },
1388
+ {
1389
+ "current_steps": 231,
1390
+ "loss": 0.6307,
1391
+ "learning_rate": 7.239130434782609e-07,
1392
+ "epoch": 2.0714285714285716
1393
+ },
1394
+ {
1395
+ "current_steps": 232,
1396
+ "loss": 0.7852,
1397
+ "learning_rate": 7.217391304347826e-07,
1398
+ "epoch": 2.080357142857143
1399
+ },
1400
+ {
1401
+ "current_steps": 233,
1402
+ "loss": 0.4894,
1403
+ "learning_rate": 7.195652173913042e-07,
1404
+ "epoch": 2.0892857142857144
1405
+ },
1406
+ {
1407
+ "current_steps": 234,
1408
+ "loss": 0.6806,
1409
+ "learning_rate": 7.17391304347826e-07,
1410
+ "epoch": 2.0982142857142856
1411
+ },
1412
+ {
1413
+ "current_steps": 235,
1414
+ "loss": 0.7798,
1415
+ "learning_rate": 7.152173913043478e-07,
1416
+ "epoch": 2.107142857142857
1417
+ },
1418
+ {
1419
+ "current_steps": 236,
1420
+ "loss": 0.934,
1421
+ "learning_rate": 7.130434782608695e-07,
1422
+ "epoch": 2.1160714285714284
1423
+ },
1424
+ {
1425
+ "current_steps": 237,
1426
+ "loss": 0.8044,
1427
+ "learning_rate": 7.108695652173913e-07,
1428
+ "epoch": 2.125
1429
+ },
1430
+ {
1431
+ "current_steps": 238,
1432
+ "loss": 0.8984,
1433
+ "learning_rate": 7.08695652173913e-07,
1434
+ "epoch": 2.1339285714285716
1435
+ },
1436
+ {
1437
+ "current_steps": 239,
1438
+ "loss": 0.7468,
1439
+ "learning_rate": 7.065217391304348e-07,
1440
+ "epoch": 2.142857142857143
1441
+ },
1442
+ {
1443
+ "current_steps": 240,
1444
+ "loss": 0.744,
1445
+ "learning_rate": 7.043478260869565e-07,
1446
+ "epoch": 2.1517857142857144
1447
+ },
1448
+ {
1449
+ "current_steps": 241,
1450
+ "loss": 0.5531,
1451
+ "learning_rate": 7.021739130434783e-07,
1452
+ "epoch": 2.1607142857142856
1453
+ },
1454
+ {
1455
+ "current_steps": 242,
1456
+ "loss": 0.8155,
1457
+ "learning_rate": 7e-07,
1458
+ "epoch": 2.169642857142857
1459
+ },
1460
+ {
1461
+ "current_steps": 243,
1462
+ "loss": 0.7626,
1463
+ "learning_rate": 6.978260869565217e-07,
1464
+ "epoch": 2.1785714285714284
1465
+ },
1466
+ {
1467
+ "current_steps": 244,
1468
+ "loss": 0.5438,
1469
+ "learning_rate": 6.956521739130434e-07,
1470
+ "epoch": 2.1875
1471
+ },
1472
+ {
1473
+ "current_steps": 245,
1474
+ "loss": 0.7638,
1475
+ "learning_rate": 6.934782608695652e-07,
1476
+ "epoch": 2.1964285714285716
1477
+ },
1478
+ {
1479
+ "current_steps": 246,
1480
+ "loss": 0.5092,
1481
+ "learning_rate": 6.913043478260869e-07,
1482
+ "epoch": 2.205357142857143
1483
+ },
1484
+ {
1485
+ "current_steps": 247,
1486
+ "loss": 0.7026,
1487
+ "learning_rate": 6.891304347826086e-07,
1488
+ "epoch": 2.2142857142857144
1489
+ },
1490
+ {
1491
+ "current_steps": 248,
1492
+ "loss": 0.727,
1493
+ "learning_rate": 6.869565217391305e-07,
1494
+ "epoch": 2.2232142857142856
1495
+ },
1496
+ {
1497
+ "current_steps": 249,
1498
+ "loss": 0.6229,
1499
+ "learning_rate": 6.847826086956522e-07,
1500
+ "epoch": 2.232142857142857
1501
+ },
1502
+ {
1503
+ "current_steps": 250,
1504
+ "loss": 0.6695,
1505
+ "learning_rate": 6.826086956521738e-07,
1506
+ "epoch": 2.2410714285714284
1507
+ },
1508
+ {
1509
+ "current_steps": 251,
1510
+ "loss": 0.6603,
1511
+ "learning_rate": 6.804347826086956e-07,
1512
+ "epoch": 2.25
1513
+ },
1514
+ {
1515
+ "current_steps": 252,
1516
+ "loss": 0.7804,
1517
+ "learning_rate": 6.782608695652173e-07,
1518
+ "epoch": 2.2589285714285716
1519
+ },
1520
+ {
1521
+ "current_steps": 253,
1522
+ "loss": 0.9138,
1523
+ "learning_rate": 6.76086956521739e-07,
1524
+ "epoch": 2.267857142857143
1525
+ },
1526
+ {
1527
+ "current_steps": 254,
1528
+ "loss": 0.7793,
1529
+ "learning_rate": 6.739130434782609e-07,
1530
+ "epoch": 2.2767857142857144
1531
+ },
1532
+ {
1533
+ "current_steps": 255,
1534
+ "loss": 0.7045,
1535
+ "learning_rate": 6.717391304347826e-07,
1536
+ "epoch": 2.2857142857142856
1537
+ },
1538
+ {
1539
+ "current_steps": 256,
1540
+ "loss": 0.8594,
1541
+ "learning_rate": 6.695652173913044e-07,
1542
+ "epoch": 2.294642857142857
1543
+ },
1544
+ {
1545
+ "current_steps": 257,
1546
+ "loss": 0.9529,
1547
+ "learning_rate": 6.673913043478261e-07,
1548
+ "epoch": 2.3035714285714284
1549
+ },
1550
+ {
1551
+ "current_steps": 258,
1552
+ "loss": 0.7477,
1553
+ "learning_rate": 6.652173913043478e-07,
1554
+ "epoch": 2.3125
1555
+ },
1556
+ {
1557
+ "current_steps": 259,
1558
+ "loss": 0.7676,
1559
+ "learning_rate": 6.630434782608695e-07,
1560
+ "epoch": 2.3214285714285716
1561
+ },
1562
+ {
1563
+ "current_steps": 260,
1564
+ "loss": 0.6468,
1565
+ "learning_rate": 6.608695652173912e-07,
1566
+ "epoch": 2.330357142857143
1567
+ },
1568
+ {
1569
+ "current_steps": 261,
1570
+ "loss": 0.6665,
1571
+ "learning_rate": 6.58695652173913e-07,
1572
+ "epoch": 2.3392857142857144
1573
+ },
1574
+ {
1575
+ "current_steps": 262,
1576
+ "loss": 0.838,
1577
+ "learning_rate": 6.565217391304348e-07,
1578
+ "epoch": 2.3482142857142856
1579
+ },
1580
+ {
1581
+ "current_steps": 263,
1582
+ "loss": 0.7129,
1583
+ "learning_rate": 6.543478260869565e-07,
1584
+ "epoch": 2.357142857142857
1585
+ },
1586
+ {
1587
+ "current_steps": 264,
1588
+ "loss": 0.8685,
1589
+ "learning_rate": 6.521739130434782e-07,
1590
+ "epoch": 2.3660714285714284
1591
+ },
1592
+ {
1593
+ "current_steps": 265,
1594
+ "loss": 0.7224,
1595
+ "learning_rate": 6.5e-07,
1596
+ "epoch": 2.375
1597
+ },
1598
+ {
1599
+ "current_steps": 266,
1600
+ "loss": 0.7037,
1601
+ "learning_rate": 6.478260869565217e-07,
1602
+ "epoch": 2.3839285714285716
1603
+ },
1604
+ {
1605
+ "current_steps": 267,
1606
+ "loss": 0.5596,
1607
+ "learning_rate": 6.456521739130435e-07,
1608
+ "epoch": 2.392857142857143
1609
+ },
1610
+ {
1611
+ "current_steps": 268,
1612
+ "loss": 0.8887,
1613
+ "learning_rate": 6.434782608695652e-07,
1614
+ "epoch": 2.4017857142857144
1615
+ },
1616
+ {
1617
+ "current_steps": 269,
1618
+ "loss": 0.6721,
1619
+ "learning_rate": 6.413043478260869e-07,
1620
+ "epoch": 2.4107142857142856
1621
+ },
1622
+ {
1623
+ "current_steps": 270,
1624
+ "loss": 0.7387,
1625
+ "learning_rate": 6.391304347826086e-07,
1626
+ "epoch": 2.419642857142857
1627
+ },
1628
+ {
1629
+ "current_steps": 271,
1630
+ "loss": 0.6304,
1631
+ "learning_rate": 6.369565217391304e-07,
1632
+ "epoch": 2.4285714285714284
1633
+ },
1634
+ {
1635
+ "current_steps": 272,
1636
+ "loss": 0.7563,
1637
+ "learning_rate": 6.347826086956521e-07,
1638
+ "epoch": 2.4375
1639
+ },
1640
+ {
1641
+ "current_steps": 273,
1642
+ "loss": 0.6833,
1643
+ "learning_rate": 6.326086956521739e-07,
1644
+ "epoch": 2.4464285714285716
1645
+ },
1646
+ {
1647
+ "current_steps": 274,
1648
+ "loss": 0.722,
1649
+ "learning_rate": 6.304347826086957e-07,
1650
+ "epoch": 2.455357142857143
1651
+ },
1652
+ {
1653
+ "current_steps": 275,
1654
+ "loss": 0.8583,
1655
+ "learning_rate": 6.282608695652174e-07,
1656
+ "epoch": 2.4642857142857144
1657
+ },
1658
+ {
1659
+ "current_steps": 276,
1660
+ "loss": 0.8988,
1661
+ "learning_rate": 6.260869565217392e-07,
1662
+ "epoch": 2.4732142857142856
1663
+ },
1664
+ {
1665
+ "current_steps": 277,
1666
+ "loss": 0.6269,
1667
+ "learning_rate": 6.239130434782608e-07,
1668
+ "epoch": 2.482142857142857
1669
+ },
1670
+ {
1671
+ "current_steps": 278,
1672
+ "loss": 0.473,
1673
+ "learning_rate": 6.217391304347825e-07,
1674
+ "epoch": 2.4910714285714284
1675
+ },
1676
+ {
1677
+ "current_steps": 279,
1678
+ "loss": 0.7065,
1679
+ "learning_rate": 6.195652173913043e-07,
1680
+ "epoch": 2.5
1681
+ },
1682
+ {
1683
+ "current_steps": 280,
1684
+ "loss": 0.7912,
1685
+ "learning_rate": 6.17391304347826e-07,
1686
+ "epoch": 2.508928571428571
1687
+ },
1688
+ {
1689
+ "current_steps": 281,
1690
+ "loss": 0.6589,
1691
+ "learning_rate": 6.152173913043478e-07,
1692
+ "epoch": 2.517857142857143
1693
+ },
1694
+ {
1695
+ "current_steps": 282,
1696
+ "loss": 0.5908,
1697
+ "learning_rate": 6.130434782608696e-07,
1698
+ "epoch": 2.5267857142857144
1699
+ },
1700
+ {
1701
+ "current_steps": 283,
1702
+ "loss": 0.839,
1703
+ "learning_rate": 6.108695652173913e-07,
1704
+ "epoch": 2.5357142857142856
1705
+ },
1706
+ {
1707
+ "current_steps": 284,
1708
+ "loss": 0.9573,
1709
+ "learning_rate": 6.08695652173913e-07,
1710
+ "epoch": 2.544642857142857
1711
+ },
1712
+ {
1713
+ "current_steps": 285,
1714
+ "loss": 0.8881,
1715
+ "learning_rate": 6.065217391304348e-07,
1716
+ "epoch": 2.553571428571429
1717
+ },
1718
+ {
1719
+ "current_steps": 286,
1720
+ "loss": 0.5213,
1721
+ "learning_rate": 6.043478260869564e-07,
1722
+ "epoch": 2.5625
1723
+ },
1724
+ {
1725
+ "current_steps": 287,
1726
+ "loss": 0.5668,
1727
+ "learning_rate": 6.021739130434782e-07,
1728
+ "epoch": 2.571428571428571
1729
+ },
1730
+ {
1731
+ "current_steps": 288,
1732
+ "loss": 0.6856,
1733
+ "learning_rate": 6e-07,
1734
+ "epoch": 2.580357142857143
1735
+ },
1736
+ {
1737
+ "current_steps": 289,
1738
+ "loss": 0.6793,
1739
+ "learning_rate": 5.978260869565217e-07,
1740
+ "epoch": 2.5892857142857144
1741
+ },
1742
+ {
1743
+ "current_steps": 290,
1744
+ "loss": 0.6176,
1745
+ "learning_rate": 5.956521739130435e-07,
1746
+ "epoch": 2.5982142857142856
1747
+ },
1748
+ {
1749
+ "current_steps": 291,
1750
+ "loss": 0.5633,
1751
+ "learning_rate": 5.934782608695652e-07,
1752
+ "epoch": 2.607142857142857
1753
+ },
1754
+ {
1755
+ "current_steps": 292,
1756
+ "loss": 0.8512,
1757
+ "learning_rate": 5.913043478260869e-07,
1758
+ "epoch": 2.616071428571429
1759
+ },
1760
+ {
1761
+ "current_steps": 293,
1762
+ "loss": 0.9664,
1763
+ "learning_rate": 5.891304347826088e-07,
1764
+ "epoch": 2.625
1765
+ },
1766
+ {
1767
+ "current_steps": 294,
1768
+ "loss": 0.6124,
1769
+ "learning_rate": 5.869565217391305e-07,
1770
+ "epoch": 2.633928571428571
1771
+ },
1772
+ {
1773
+ "current_steps": 295,
1774
+ "loss": 0.6244,
1775
+ "learning_rate": 5.847826086956521e-07,
1776
+ "epoch": 2.642857142857143
1777
+ },
1778
+ {
1779
+ "current_steps": 296,
1780
+ "loss": 0.7879,
1781
+ "learning_rate": 5.826086956521739e-07,
1782
+ "epoch": 2.6517857142857144
1783
+ },
1784
+ {
1785
+ "current_steps": 297,
1786
+ "loss": 0.6862,
1787
+ "learning_rate": 5.804347826086956e-07,
1788
+ "epoch": 2.6607142857142856
1789
+ },
1790
+ {
1791
+ "current_steps": 298,
1792
+ "loss": 0.6368,
1793
+ "learning_rate": 5.782608695652173e-07,
1794
+ "epoch": 2.669642857142857
1795
+ },
1796
+ {
1797
+ "current_steps": 299,
1798
+ "loss": 0.8478,
1799
+ "learning_rate": 5.760869565217391e-07,
1800
+ "epoch": 2.678571428571429
1801
+ },
1802
+ {
1803
+ "current_steps": 300,
1804
+ "loss": 0.6466,
1805
+ "learning_rate": 5.739130434782609e-07,
1806
+ "epoch": 2.6875
1807
+ },
1808
+ {
1809
+ "current_steps": 301,
1810
+ "loss": 0.7323,
1811
+ "learning_rate": 5.717391304347826e-07,
1812
+ "epoch": 2.696428571428571
1813
+ },
1814
+ {
1815
+ "current_steps": 302,
1816
+ "loss": 0.7611,
1817
+ "learning_rate": 5.695652173913044e-07,
1818
+ "epoch": 2.705357142857143
1819
+ },
1820
+ {
1821
+ "current_steps": 303,
1822
+ "loss": 0.7075,
1823
+ "learning_rate": 5.673913043478261e-07,
1824
+ "epoch": 2.7142857142857144
1825
+ },
1826
+ {
1827
+ "current_steps": 304,
1828
+ "loss": 0.5448,
1829
+ "learning_rate": 5.652173913043477e-07,
1830
+ "epoch": 2.7232142857142856
1831
+ },
1832
+ {
1833
+ "current_steps": 305,
1834
+ "loss": 0.704,
1835
+ "learning_rate": 5.630434782608695e-07,
1836
+ "epoch": 2.732142857142857
1837
+ },
1838
+ {
1839
+ "current_steps": 306,
1840
+ "loss": 0.8591,
1841
+ "learning_rate": 5.608695652173912e-07,
1842
+ "epoch": 2.741071428571429
1843
+ },
1844
+ {
1845
+ "current_steps": 307,
1846
+ "loss": 0.6702,
1847
+ "learning_rate": 5.58695652173913e-07,
1848
+ "epoch": 2.75
1849
+ },
1850
+ {
1851
+ "current_steps": 308,
1852
+ "loss": 0.6652,
1853
+ "learning_rate": 5.565217391304348e-07,
1854
+ "epoch": 2.758928571428571
1855
+ },
1856
+ {
1857
+ "current_steps": 309,
1858
+ "loss": 0.7208,
1859
+ "learning_rate": 5.543478260869565e-07,
1860
+ "epoch": 2.767857142857143
1861
+ },
1862
+ {
1863
+ "current_steps": 310,
1864
+ "loss": 0.7334,
1865
+ "learning_rate": 5.521739130434783e-07,
1866
+ "epoch": 2.7767857142857144
1867
+ },
1868
+ {
1869
+ "current_steps": 311,
1870
+ "loss": 0.865,
1871
+ "learning_rate": 5.5e-07,
1872
+ "epoch": 2.7857142857142856
1873
+ },
1874
+ {
1875
+ "current_steps": 312,
1876
+ "loss": 0.5955,
1877
+ "learning_rate": 5.478260869565216e-07,
1878
+ "epoch": 2.794642857142857
1879
+ },
1880
+ {
1881
+ "current_steps": 313,
1882
+ "loss": 0.5059,
1883
+ "learning_rate": 5.456521739130435e-07,
1884
+ "epoch": 2.803571428571429
1885
+ },
1886
+ {
1887
+ "current_steps": 314,
1888
+ "loss": 1.0855,
1889
+ "learning_rate": 5.434782608695652e-07,
1890
+ "epoch": 2.8125
1891
+ },
1892
+ {
1893
+ "current_steps": 315,
1894
+ "loss": 0.7484,
1895
+ "learning_rate": 5.413043478260869e-07,
1896
+ "epoch": 2.821428571428571
1897
+ },
1898
+ {
1899
+ "current_steps": 316,
1900
+ "loss": 0.8017,
1901
+ "learning_rate": 5.391304347826087e-07,
1902
+ "epoch": 2.830357142857143
1903
+ },
1904
+ {
1905
+ "current_steps": 317,
1906
+ "loss": 0.7272,
1907
+ "learning_rate": 5.369565217391304e-07,
1908
+ "epoch": 2.8392857142857144
1909
+ },
1910
+ {
1911
+ "current_steps": 318,
1912
+ "loss": 0.6897,
1913
+ "learning_rate": 5.347826086956521e-07,
1914
+ "epoch": 2.8482142857142856
1915
+ },
1916
+ {
1917
+ "current_steps": 319,
1918
+ "loss": 0.634,
1919
+ "learning_rate": 5.32608695652174e-07,
1920
+ "epoch": 2.857142857142857
1921
+ },
1922
+ {
1923
+ "current_steps": 320,
1924
+ "loss": 0.7684,
1925
+ "learning_rate": 5.304347826086957e-07,
1926
+ "epoch": 2.866071428571429
1927
+ },
1928
+ {
1929
+ "current_steps": 321,
1930
+ "loss": 0.5758,
1931
+ "learning_rate": 5.282608695652173e-07,
1932
+ "epoch": 2.875
1933
+ },
1934
+ {
1935
+ "current_steps": 322,
1936
+ "loss": 0.687,
1937
+ "learning_rate": 5.260869565217391e-07,
1938
+ "epoch": 2.883928571428571
1939
+ },
1940
+ {
1941
+ "current_steps": 323,
1942
+ "loss": 0.6942,
1943
+ "learning_rate": 5.239130434782608e-07,
1944
+ "epoch": 2.892857142857143
1945
+ },
1946
+ {
1947
+ "current_steps": 324,
1948
+ "loss": 0.7698,
1949
+ "learning_rate": 5.217391304347825e-07,
1950
+ "epoch": 2.9017857142857144
1951
+ },
1952
+ {
1953
+ "current_steps": 325,
1954
+ "loss": 0.815,
1955
+ "learning_rate": 5.195652173913043e-07,
1956
+ "epoch": 2.9107142857142856
1957
+ },
1958
+ {
1959
+ "current_steps": 326,
1960
+ "loss": 0.6837,
1961
+ "learning_rate": 5.173913043478261e-07,
1962
+ "epoch": 2.919642857142857
1963
+ },
1964
+ {
1965
+ "current_steps": 327,
1966
+ "loss": 0.7103,
1967
+ "learning_rate": 5.152173913043479e-07,
1968
+ "epoch": 2.928571428571429
1969
+ },
1970
+ {
1971
+ "current_steps": 328,
1972
+ "loss": 0.6798,
1973
+ "learning_rate": 5.130434782608696e-07,
1974
+ "epoch": 2.9375
1975
+ },
1976
+ {
1977
+ "current_steps": 329,
1978
+ "loss": 0.767,
1979
+ "learning_rate": 5.108695652173913e-07,
1980
+ "epoch": 2.946428571428571
1981
+ },
1982
+ {
1983
+ "current_steps": 330,
1984
+ "loss": 0.6161,
1985
+ "learning_rate": 5.08695652173913e-07,
1986
+ "epoch": 2.955357142857143
1987
+ },
1988
+ {
1989
+ "current_steps": 331,
1990
+ "loss": 0.6607,
1991
+ "learning_rate": 5.065217391304347e-07,
1992
+ "epoch": 2.9642857142857144
1993
+ },
1994
+ {
1995
+ "current_steps": 332,
1996
+ "loss": 0.6875,
1997
+ "learning_rate": 5.043478260869564e-07,
1998
+ "epoch": 2.9732142857142856
1999
+ },
2000
+ {
2001
+ "current_steps": 333,
2002
+ "loss": 0.746,
2003
+ "learning_rate": 5.021739130434783e-07,
2004
+ "epoch": 2.982142857142857
2005
+ },
2006
+ {
2007
+ "current_steps": 334,
2008
+ "loss": 0.6093,
2009
+ "learning_rate": 5e-07,
2010
+ "epoch": 2.991071428571429
2011
+ },
2012
+ {
2013
+ "current_steps": 335,
2014
+ "loss": 0.5599,
2015
+ "learning_rate": 4.978260869565217e-07,
2016
+ "epoch": 3.0
2017
+ },
2018
+ {
2019
+ "current_steps": 336,
2020
+ "loss": 0.5985,
2021
+ "learning_rate": 4.956521739130435e-07,
2022
+ "epoch": 3.0089285714285716
2023
+ },
2024
+ {
2025
+ "current_steps": 337,
2026
+ "loss": 0.6692,
2027
+ "learning_rate": 4.934782608695652e-07,
2028
+ "epoch": 3.017857142857143
2029
+ },
2030
+ {
2031
+ "current_steps": 338,
2032
+ "loss": 0.5887,
2033
+ "learning_rate": 4.913043478260869e-07,
2034
+ "epoch": 3.0267857142857144
2035
+ },
2036
+ {
2037
+ "current_steps": 339,
2038
+ "loss": 0.5831,
2039
+ "learning_rate": 4.891304347826087e-07,
2040
+ "epoch": 3.0357142857142856
2041
+ },
2042
+ {
2043
+ "current_steps": 340,
2044
+ "loss": 0.5424,
2045
+ "learning_rate": 4.869565217391305e-07,
2046
+ "epoch": 3.044642857142857
2047
+ },
2048
+ {
2049
+ "current_steps": 341,
2050
+ "loss": 1.0041,
2051
+ "learning_rate": 4.847826086956521e-07,
2052
+ "epoch": 3.0535714285714284
2053
+ },
2054
+ {
2055
+ "current_steps": 342,
2056
+ "loss": 0.6989,
2057
+ "learning_rate": 4.826086956521739e-07,
2058
+ "epoch": 3.0625
2059
+ },
2060
+ {
2061
+ "current_steps": 343,
2062
+ "loss": 0.7104,
2063
+ "learning_rate": 4.804347826086956e-07,
2064
+ "epoch": 3.0714285714285716
2065
+ },
2066
+ {
2067
+ "current_steps": 344,
2068
+ "loss": 0.6493,
2069
+ "learning_rate": 4.782608695652174e-07,
2070
+ "epoch": 3.080357142857143
2071
+ },
2072
+ {
2073
+ "current_steps": 345,
2074
+ "loss": 0.8018,
2075
+ "learning_rate": 4.7608695652173915e-07,
2076
+ "epoch": 3.0892857142857144
2077
+ },
2078
+ {
2079
+ "current_steps": 346,
2080
+ "loss": 0.638,
2081
+ "learning_rate": 4.739130434782608e-07,
2082
+ "epoch": 3.0982142857142856
2083
+ },
2084
+ {
2085
+ "current_steps": 347,
2086
+ "loss": 0.7714,
2087
+ "learning_rate": 4.717391304347826e-07,
2088
+ "epoch": 3.107142857142857
2089
+ },
2090
+ {
2091
+ "current_steps": 348,
2092
+ "loss": 0.7103,
2093
+ "learning_rate": 4.6956521739130434e-07,
2094
+ "epoch": 3.1160714285714284
2095
+ },
2096
+ {
2097
+ "current_steps": 349,
2098
+ "loss": 0.5937,
2099
+ "learning_rate": 4.673913043478261e-07,
2100
+ "epoch": 3.125
2101
+ },
2102
+ {
2103
+ "current_steps": 350,
2104
+ "loss": 0.7256,
2105
+ "learning_rate": 4.6521739130434777e-07,
2106
+ "epoch": 3.1339285714285716
2107
+ },
2108
+ {
2109
+ "current_steps": 351,
2110
+ "loss": 0.864,
2111
+ "learning_rate": 4.6304347826086954e-07,
2112
+ "epoch": 3.142857142857143
2113
+ },
2114
+ {
2115
+ "current_steps": 352,
2116
+ "loss": 0.7429,
2117
+ "learning_rate": 4.608695652173913e-07,
2118
+ "epoch": 3.1517857142857144
2119
+ },
2120
+ {
2121
+ "current_steps": 353,
2122
+ "loss": 0.6658,
2123
+ "learning_rate": 4.58695652173913e-07,
2124
+ "epoch": 3.1607142857142856
2125
+ },
2126
+ {
2127
+ "current_steps": 354,
2128
+ "loss": 0.647,
2129
+ "learning_rate": 4.5652173913043473e-07,
2130
+ "epoch": 3.169642857142857
2131
+ },
2132
+ {
2133
+ "current_steps": 355,
2134
+ "loss": 0.7772,
2135
+ "learning_rate": 4.543478260869565e-07,
2136
+ "epoch": 3.1785714285714284
2137
+ },
2138
+ {
2139
+ "current_steps": 356,
2140
+ "loss": 0.6939,
2141
+ "learning_rate": 4.521739130434782e-07,
2142
+ "epoch": 3.1875
2143
+ },
2144
+ {
2145
+ "current_steps": 357,
2146
+ "loss": 0.5744,
2147
+ "learning_rate": 4.5e-07,
2148
+ "epoch": 3.1964285714285716
2149
+ },
2150
+ {
2151
+ "current_steps": 358,
2152
+ "loss": 0.7193,
2153
+ "learning_rate": 4.4782608695652175e-07,
2154
+ "epoch": 3.205357142857143
2155
+ },
2156
+ {
2157
+ "current_steps": 359,
2158
+ "loss": 0.667,
2159
+ "learning_rate": 4.4565217391304346e-07,
2160
+ "epoch": 3.2142857142857144
2161
+ },
2162
+ {
2163
+ "current_steps": 360,
2164
+ "loss": 0.6671,
2165
+ "learning_rate": 4.434782608695652e-07,
2166
+ "epoch": 3.2232142857142856
2167
+ },
2168
+ {
2169
+ "current_steps": 361,
2170
+ "loss": 0.8531,
2171
+ "learning_rate": 4.4130434782608694e-07,
2172
+ "epoch": 3.232142857142857
2173
+ },
2174
+ {
2175
+ "current_steps": 362,
2176
+ "loss": 0.6706,
2177
+ "learning_rate": 4.391304347826087e-07,
2178
+ "epoch": 3.2410714285714284
2179
+ },
2180
+ {
2181
+ "current_steps": 363,
2182
+ "loss": 0.8786,
2183
+ "learning_rate": 4.3695652173913037e-07,
2184
+ "epoch": 3.25
2185
+ },
2186
+ {
2187
+ "current_steps": 364,
2188
+ "loss": 0.6281,
2189
+ "learning_rate": 4.3478260869565214e-07,
2190
+ "epoch": 3.2589285714285716
2191
+ },
2192
+ {
2193
+ "current_steps": 365,
2194
+ "loss": 0.8648,
2195
+ "learning_rate": 4.326086956521739e-07,
2196
+ "epoch": 3.267857142857143
2197
+ },
2198
+ {
2199
+ "current_steps": 366,
2200
+ "loss": 0.5872,
2201
+ "learning_rate": 4.3043478260869567e-07,
2202
+ "epoch": 3.2767857142857144
2203
+ },
2204
+ {
2205
+ "current_steps": 367,
2206
+ "loss": 0.5874,
2207
+ "learning_rate": 4.282608695652174e-07,
2208
+ "epoch": 3.2857142857142856
2209
+ },
2210
+ {
2211
+ "current_steps": 368,
2212
+ "loss": 0.7057,
2213
+ "learning_rate": 4.260869565217391e-07,
2214
+ "epoch": 3.294642857142857
2215
+ },
2216
+ {
2217
+ "current_steps": 369,
2218
+ "loss": 0.6076,
2219
+ "learning_rate": 4.2391304347826086e-07,
2220
+ "epoch": 3.3035714285714284
2221
+ },
2222
+ {
2223
+ "current_steps": 370,
2224
+ "loss": 0.7514,
2225
+ "learning_rate": 4.217391304347826e-07,
2226
+ "epoch": 3.3125
2227
+ },
2228
+ {
2229
+ "current_steps": 371,
2230
+ "loss": 0.689,
2231
+ "learning_rate": 4.1956521739130434e-07,
2232
+ "epoch": 3.3214285714285716
2233
+ },
2234
+ {
2235
+ "current_steps": 372,
2236
+ "loss": 0.7074,
2237
+ "learning_rate": 4.1739130434782606e-07,
2238
+ "epoch": 3.330357142857143
2239
+ },
2240
+ {
2241
+ "current_steps": 373,
2242
+ "loss": 0.6425,
2243
+ "learning_rate": 4.1521739130434777e-07,
2244
+ "epoch": 3.3392857142857144
2245
+ },
2246
+ {
2247
+ "current_steps": 374,
2248
+ "loss": 0.5247,
2249
+ "learning_rate": 4.1304347826086954e-07,
2250
+ "epoch": 3.3482142857142856
2251
+ },
2252
+ {
2253
+ "current_steps": 375,
2254
+ "loss": 0.7755,
2255
+ "learning_rate": 4.108695652173913e-07,
2256
+ "epoch": 3.357142857142857
2257
+ },
2258
+ {
2259
+ "current_steps": 376,
2260
+ "loss": 0.7774,
2261
+ "learning_rate": 4.0869565217391307e-07,
2262
+ "epoch": 3.3660714285714284
2263
+ },
2264
+ {
2265
+ "current_steps": 377,
2266
+ "loss": 0.6871,
2267
+ "learning_rate": 4.0652173913043473e-07,
2268
+ "epoch": 3.375
2269
+ },
2270
+ {
2271
+ "current_steps": 378,
2272
+ "loss": 0.566,
2273
+ "learning_rate": 4.043478260869565e-07,
2274
+ "epoch": 3.3839285714285716
2275
+ },
2276
+ {
2277
+ "current_steps": 379,
2278
+ "loss": 1.0922,
2279
+ "learning_rate": 4.0217391304347827e-07,
2280
+ "epoch": 3.392857142857143
2281
+ },
2282
+ {
2283
+ "current_steps": 380,
2284
+ "loss": 0.5958,
2285
+ "learning_rate": 4e-07,
2286
+ "epoch": 3.4017857142857144
2287
+ },
2288
+ {
2289
+ "current_steps": 381,
2290
+ "loss": 0.9182,
2291
+ "learning_rate": 3.978260869565217e-07,
2292
+ "epoch": 3.4107142857142856
2293
+ },
2294
+ {
2295
+ "current_steps": 382,
2296
+ "loss": 0.7356,
2297
+ "learning_rate": 3.9565217391304346e-07,
2298
+ "epoch": 3.419642857142857
2299
+ },
2300
+ {
2301
+ "current_steps": 383,
2302
+ "loss": 0.8677,
2303
+ "learning_rate": 3.9347826086956523e-07,
2304
+ "epoch": 3.4285714285714284
2305
+ },
2306
+ {
2307
+ "current_steps": 384,
2308
+ "loss": 0.6885,
2309
+ "learning_rate": 3.9130434782608694e-07,
2310
+ "epoch": 3.4375
2311
+ },
2312
+ {
2313
+ "current_steps": 385,
2314
+ "loss": 0.7982,
2315
+ "learning_rate": 3.8913043478260866e-07,
2316
+ "epoch": 3.4464285714285716
2317
+ },
2318
+ {
2319
+ "current_steps": 386,
2320
+ "loss": 0.8466,
2321
+ "learning_rate": 3.869565217391304e-07,
2322
+ "epoch": 3.455357142857143
2323
+ },
2324
+ {
2325
+ "current_steps": 387,
2326
+ "loss": 0.4563,
2327
+ "learning_rate": 3.8478260869565214e-07,
2328
+ "epoch": 3.4642857142857144
2329
+ },
2330
+ {
2331
+ "current_steps": 388,
2332
+ "loss": 0.7675,
2333
+ "learning_rate": 3.826086956521739e-07,
2334
+ "epoch": 3.4732142857142856
2335
+ },
2336
+ {
2337
+ "current_steps": 389,
2338
+ "loss": 0.7642,
2339
+ "learning_rate": 3.8043478260869567e-07,
2340
+ "epoch": 3.482142857142857
2341
+ },
2342
+ {
2343
+ "current_steps": 390,
2344
+ "loss": 0.6065,
2345
+ "learning_rate": 3.7826086956521733e-07,
2346
+ "epoch": 3.4910714285714284
2347
+ },
2348
+ {
2349
+ "current_steps": 391,
2350
+ "loss": 0.6121,
2351
+ "learning_rate": 3.760869565217391e-07,
2352
+ "epoch": 3.5
2353
+ },
2354
+ {
2355
+ "current_steps": 392,
2356
+ "loss": 0.8562,
2357
+ "learning_rate": 3.7391304347826087e-07,
2358
+ "epoch": 3.508928571428571
2359
+ },
2360
+ {
2361
+ "current_steps": 393,
2362
+ "loss": 0.8169,
2363
+ "learning_rate": 3.7173913043478263e-07,
2364
+ "epoch": 3.517857142857143
2365
+ },
2366
+ {
2367
+ "current_steps": 394,
2368
+ "loss": 0.7264,
2369
+ "learning_rate": 3.695652173913043e-07,
2370
+ "epoch": 3.5267857142857144
2371
+ },
2372
+ {
2373
+ "current_steps": 395,
2374
+ "loss": 0.6761,
2375
+ "learning_rate": 3.6739130434782606e-07,
2376
+ "epoch": 3.5357142857142856
2377
+ },
2378
+ {
2379
+ "current_steps": 396,
2380
+ "loss": 0.485,
2381
+ "learning_rate": 3.6521739130434783e-07,
2382
+ "epoch": 3.544642857142857
2383
+ },
2384
+ {
2385
+ "current_steps": 397,
2386
+ "loss": 0.6992,
2387
+ "learning_rate": 3.6304347826086954e-07,
2388
+ "epoch": 3.553571428571429
2389
+ },
2390
+ {
2391
+ "current_steps": 398,
2392
+ "loss": 0.6543,
2393
+ "learning_rate": 3.608695652173913e-07,
2394
+ "epoch": 3.5625
2395
+ },
2396
+ {
2397
+ "current_steps": 399,
2398
+ "loss": 0.6019,
2399
+ "learning_rate": 3.58695652173913e-07,
2400
+ "epoch": 3.571428571428571
2401
+ },
2402
+ {
2403
+ "current_steps": 400,
2404
+ "loss": 0.8135,
2405
+ "learning_rate": 3.5652173913043474e-07,
2406
+ "epoch": 3.580357142857143
2407
+ },
2408
+ {
2409
+ "current_steps": 401,
2410
+ "loss": 0.5053,
2411
+ "learning_rate": 3.543478260869565e-07,
2412
+ "epoch": 3.5892857142857144
2413
+ },
2414
+ {
2415
+ "current_steps": 402,
2416
+ "loss": 0.6121,
2417
+ "learning_rate": 3.5217391304347827e-07,
2418
+ "epoch": 3.5982142857142856
2419
+ },
2420
+ {
2421
+ "current_steps": 403,
2422
+ "loss": 0.5648,
2423
+ "learning_rate": 3.5e-07,
2424
+ "epoch": 3.607142857142857
2425
+ },
2426
+ {
2427
+ "current_steps": 404,
2428
+ "loss": 0.6023,
2429
+ "learning_rate": 3.478260869565217e-07,
2430
+ "epoch": 3.616071428571429
2431
+ },
2432
+ {
2433
+ "current_steps": 405,
2434
+ "loss": 0.7843,
2435
+ "learning_rate": 3.4565217391304346e-07,
2436
+ "epoch": 3.625
2437
+ },
2438
+ {
2439
+ "current_steps": 406,
2440
+ "loss": 0.6902,
2441
+ "learning_rate": 3.4347826086956523e-07,
2442
+ "epoch": 3.633928571428571
2443
+ },
2444
+ {
2445
+ "current_steps": 407,
2446
+ "loss": 0.6103,
2447
+ "learning_rate": 3.413043478260869e-07,
2448
+ "epoch": 3.642857142857143
2449
+ },
2450
+ {
2451
+ "current_steps": 408,
2452
+ "loss": 0.759,
2453
+ "learning_rate": 3.3913043478260866e-07,
2454
+ "epoch": 3.6517857142857144
2455
+ },
2456
+ {
2457
+ "current_steps": 409,
2458
+ "loss": 0.7823,
2459
+ "learning_rate": 3.369565217391304e-07,
2460
+ "epoch": 3.6607142857142856
2461
+ },
2462
+ {
2463
+ "current_steps": 410,
2464
+ "loss": 0.8021,
2465
+ "learning_rate": 3.347826086956522e-07,
2466
+ "epoch": 3.669642857142857
2467
+ },
2468
+ {
2469
+ "current_steps": 411,
2470
+ "loss": 0.5927,
2471
+ "learning_rate": 3.326086956521739e-07,
2472
+ "epoch": 3.678571428571429
2473
+ },
2474
+ {
2475
+ "current_steps": 412,
2476
+ "loss": 0.6503,
2477
+ "learning_rate": 3.304347826086956e-07,
2478
+ "epoch": 3.6875
2479
+ },
2480
+ {
2481
+ "current_steps": 413,
2482
+ "loss": 0.886,
2483
+ "learning_rate": 3.282608695652174e-07,
2484
+ "epoch": 3.696428571428571
2485
+ },
2486
+ {
2487
+ "current_steps": 414,
2488
+ "loss": 0.6331,
2489
+ "learning_rate": 3.260869565217391e-07,
2490
+ "epoch": 3.705357142857143
2491
+ },
2492
+ {
2493
+ "current_steps": 415,
2494
+ "loss": 0.7633,
2495
+ "learning_rate": 3.2391304347826087e-07,
2496
+ "epoch": 3.7142857142857144
2497
+ },
2498
+ {
2499
+ "current_steps": 416,
2500
+ "loss": 0.6538,
2501
+ "learning_rate": 3.217391304347826e-07,
2502
+ "epoch": 3.7232142857142856
2503
+ },
2504
+ {
2505
+ "current_steps": 417,
2506
+ "loss": 0.6156,
2507
+ "learning_rate": 3.195652173913043e-07,
2508
+ "epoch": 3.732142857142857
2509
+ },
2510
+ {
2511
+ "current_steps": 418,
2512
+ "loss": 0.6973,
2513
+ "learning_rate": 3.1739130434782606e-07,
2514
+ "epoch": 3.741071428571429
2515
+ },
2516
+ {
2517
+ "current_steps": 419,
2518
+ "loss": 0.6521,
2519
+ "learning_rate": 3.1521739130434783e-07,
2520
+ "epoch": 3.75
2521
+ },
2522
+ {
2523
+ "current_steps": 420,
2524
+ "loss": 0.6931,
2525
+ "learning_rate": 3.130434782608696e-07,
2526
+ "epoch": 3.758928571428571
2527
+ },
2528
+ {
2529
+ "current_steps": 421,
2530
+ "loss": 0.8192,
2531
+ "learning_rate": 3.1086956521739126e-07,
2532
+ "epoch": 3.767857142857143
2533
+ },
2534
+ {
2535
+ "current_steps": 422,
2536
+ "loss": 0.5986,
2537
+ "learning_rate": 3.08695652173913e-07,
2538
+ "epoch": 3.7767857142857144
2539
+ },
2540
+ {
2541
+ "current_steps": 423,
2542
+ "loss": 0.9986,
2543
+ "learning_rate": 3.065217391304348e-07,
2544
+ "epoch": 3.7857142857142856
2545
+ },
2546
+ {
2547
+ "current_steps": 424,
2548
+ "loss": 0.7645,
2549
+ "learning_rate": 3.043478260869565e-07,
2550
+ "epoch": 3.794642857142857
2551
+ },
2552
+ {
2553
+ "current_steps": 425,
2554
+ "loss": 0.6489,
2555
+ "learning_rate": 3.021739130434782e-07,
2556
+ "epoch": 3.803571428571429
2557
+ },
2558
+ {
2559
+ "current_steps": 426,
2560
+ "loss": 0.5974,
2561
+ "learning_rate": 3e-07,
2562
+ "epoch": 3.8125
2563
+ },
2564
+ {
2565
+ "current_steps": 427,
2566
+ "loss": 0.7392,
2567
+ "learning_rate": 2.9782608695652175e-07,
2568
+ "epoch": 3.821428571428571
2569
+ },
2570
+ {
2571
+ "current_steps": 428,
2572
+ "loss": 0.7813,
2573
+ "learning_rate": 2.9565217391304347e-07,
2574
+ "epoch": 3.830357142857143
2575
+ },
2576
+ {
2577
+ "current_steps": 429,
2578
+ "loss": 0.7818,
2579
+ "learning_rate": 2.9347826086956523e-07,
2580
+ "epoch": 3.8392857142857144
2581
+ },
2582
+ {
2583
+ "current_steps": 430,
2584
+ "loss": 1.0693,
2585
+ "learning_rate": 2.9130434782608695e-07,
2586
+ "epoch": 3.8482142857142856
2587
+ },
2588
+ {
2589
+ "current_steps": 431,
2590
+ "loss": 0.6324,
2591
+ "learning_rate": 2.8913043478260866e-07,
2592
+ "epoch": 3.857142857142857
2593
+ },
2594
+ {
2595
+ "current_steps": 432,
2596
+ "loss": 0.5228,
2597
+ "learning_rate": 2.8695652173913043e-07,
2598
+ "epoch": 3.866071428571429
2599
+ },
2600
+ {
2601
+ "current_steps": 433,
2602
+ "loss": 0.6631,
2603
+ "learning_rate": 2.847826086956522e-07,
2604
+ "epoch": 3.875
2605
+ },
2606
+ {
2607
+ "current_steps": 434,
2608
+ "loss": 0.6685,
2609
+ "learning_rate": 2.8260869565217386e-07,
2610
+ "epoch": 3.883928571428571
2611
+ },
2612
+ {
2613
+ "current_steps": 435,
2614
+ "loss": 0.6566,
2615
+ "learning_rate": 2.804347826086956e-07,
2616
+ "epoch": 3.892857142857143
2617
+ },
2618
+ {
2619
+ "current_steps": 436,
2620
+ "loss": 0.6169,
2621
+ "learning_rate": 2.782608695652174e-07,
2622
+ "epoch": 3.9017857142857144
2623
+ },
2624
+ {
2625
+ "current_steps": 437,
2626
+ "loss": 0.5012,
2627
+ "learning_rate": 2.7608695652173916e-07,
2628
+ "epoch": 3.9107142857142856
2629
+ },
2630
+ {
2631
+ "current_steps": 438,
2632
+ "loss": 0.637,
2633
+ "learning_rate": 2.739130434782608e-07,
2634
+ "epoch": 3.919642857142857
2635
+ },
2636
+ {
2637
+ "current_steps": 439,
2638
+ "loss": 0.7777,
2639
+ "learning_rate": 2.717391304347826e-07,
2640
+ "epoch": 3.928571428571429
2641
+ },
2642
+ {
2643
+ "current_steps": 440,
2644
+ "loss": 0.6963,
2645
+ "learning_rate": 2.6956521739130435e-07,
2646
+ "epoch": 3.9375
2647
+ },
2648
+ {
2649
+ "current_steps": 441,
2650
+ "loss": 0.5398,
2651
+ "learning_rate": 2.6739130434782607e-07,
2652
+ "epoch": 3.946428571428571
2653
+ },
2654
+ {
2655
+ "current_steps": 442,
2656
+ "loss": 1.0029,
2657
+ "learning_rate": 2.6521739130434783e-07,
2658
+ "epoch": 3.955357142857143
2659
+ },
2660
+ {
2661
+ "current_steps": 443,
2662
+ "loss": 0.8166,
2663
+ "learning_rate": 2.6304347826086955e-07,
2664
+ "epoch": 3.9642857142857144
2665
+ },
2666
+ {
2667
+ "current_steps": 444,
2668
+ "loss": 0.8981,
2669
+ "learning_rate": 2.6086956521739126e-07,
2670
+ "epoch": 3.9732142857142856
2671
+ },
2672
+ {
2673
+ "current_steps": 445,
2674
+ "loss": 0.536,
2675
+ "learning_rate": 2.5869565217391303e-07,
2676
+ "epoch": 3.982142857142857
2677
+ },
2678
+ {
2679
+ "current_steps": 446,
2680
+ "loss": 0.7719,
2681
+ "learning_rate": 2.565217391304348e-07,
2682
+ "epoch": 3.991071428571429
2683
+ },
2684
+ {
2685
+ "current_steps": 447,
2686
+ "loss": 3.9574,
2687
+ "learning_rate": 2.565217391304348e-07,
2688
+ "epoch": 4.0
2689
+ },
2690
+ {
2691
+ "current_steps": 448,
2692
+ "loss": 0.6567,
2693
+ "learning_rate": 2.543478260869565e-07,
2694
+ "epoch": 4.008928571428571
2695
+ },
2696
+ {
2697
+ "current_steps": 449,
2698
+ "loss": 0.8622,
2699
+ "learning_rate": 2.521739130434782e-07,
2700
+ "epoch": 4.017857142857143
2701
+ },
2702
+ {
2703
+ "current_steps": 450,
2704
+ "loss": 0.5737,
2705
+ "learning_rate": 2.5e-07,
2706
+ "epoch": 4.026785714285714
2707
+ },
2708
+ {
2709
+ "current_steps": 451,
2710
+ "loss": 0.736,
2711
+ "learning_rate": 2.4782608695652176e-07,
2712
+ "epoch": 4.035714285714286
2713
+ },
2714
+ {
2715
+ "current_steps": 452,
2716
+ "loss": 0.8457,
2717
+ "learning_rate": 2.4565217391304347e-07,
2718
+ "epoch": 4.044642857142857
2719
+ },
2720
+ {
2721
+ "current_steps": 453,
2722
+ "loss": 0.7416,
2723
+ "learning_rate": 2.4347826086956524e-07,
2724
+ "epoch": 4.053571428571429
2725
+ },
2726
+ {
2727
+ "current_steps": 454,
2728
+ "loss": 1.0355,
2729
+ "learning_rate": 2.4130434782608695e-07,
2730
+ "epoch": 4.0625
2731
+ },
2732
+ {
2733
+ "current_steps": 455,
2734
+ "loss": 0.7162,
2735
+ "learning_rate": 2.391304347826087e-07,
2736
+ "epoch": 4.071428571428571
2737
+ },
2738
+ {
2739
+ "current_steps": 456,
2740
+ "loss": 0.8163,
2741
+ "learning_rate": 2.369565217391304e-07,
2742
+ "epoch": 4.080357142857143
2743
+ },
2744
+ {
2745
+ "current_steps": 457,
2746
+ "loss": 0.5188,
2747
+ "learning_rate": 2.3478260869565217e-07,
2748
+ "epoch": 4.089285714285714
2749
+ },
2750
+ {
2751
+ "current_steps": 458,
2752
+ "loss": 0.9544,
2753
+ "learning_rate": 2.3260869565217389e-07,
2754
+ "epoch": 4.098214285714286
2755
+ },
2756
+ {
2757
+ "current_steps": 459,
2758
+ "loss": 0.6205,
2759
+ "learning_rate": 2.3043478260869565e-07,
2760
+ "epoch": 4.107142857142857
2761
+ },
2762
+ {
2763
+ "current_steps": 460,
2764
+ "loss": 0.6643,
2765
+ "learning_rate": 2.2826086956521737e-07,
2766
+ "epoch": 4.116071428571429
2767
+ },
2768
+ {
2769
+ "current_steps": 461,
2770
+ "loss": 0.6465,
2771
+ "learning_rate": 2.260869565217391e-07,
2772
+ "epoch": 4.125
2773
+ },
2774
+ {
2775
+ "current_steps": 462,
2776
+ "loss": 0.6697,
2777
+ "learning_rate": 2.2391304347826087e-07,
2778
+ "epoch": 4.133928571428571
2779
+ },
2780
+ {
2781
+ "current_steps": 463,
2782
+ "loss": 0.7041,
2783
+ "learning_rate": 2.217391304347826e-07,
2784
+ "epoch": 4.142857142857143
2785
+ },
2786
+ {
2787
+ "current_steps": 464,
2788
+ "loss": 0.802,
2789
+ "learning_rate": 2.1956521739130435e-07,
2790
+ "epoch": 4.151785714285714
2791
+ },
2792
+ {
2793
+ "current_steps": 465,
2794
+ "loss": 0.623,
2795
+ "learning_rate": 2.1739130434782607e-07,
2796
+ "epoch": 4.160714285714286
2797
+ },
2798
+ {
2799
+ "current_steps": 466,
2800
+ "loss": 0.6071,
2801
+ "learning_rate": 2.1521739130434783e-07,
2802
+ "epoch": 4.169642857142857
2803
+ },
2804
+ {
2805
+ "current_steps": 467,
2806
+ "loss": 0.718,
2807
+ "learning_rate": 2.1304347826086955e-07,
2808
+ "epoch": 4.178571428571429
2809
+ },
2810
+ {
2811
+ "current_steps": 468,
2812
+ "loss": 0.6337,
2813
+ "learning_rate": 2.108695652173913e-07,
2814
+ "epoch": 4.1875
2815
+ },
2816
+ {
2817
+ "current_steps": 469,
2818
+ "loss": 0.5689,
2819
+ "learning_rate": 2.0869565217391303e-07,
2820
+ "epoch": 4.196428571428571
2821
+ },
2822
+ {
2823
+ "current_steps": 470,
2824
+ "loss": 0.62,
2825
+ "learning_rate": 2.0652173913043477e-07,
2826
+ "epoch": 4.205357142857143
2827
+ },
2828
+ {
2829
+ "current_steps": 471,
2830
+ "loss": 1.0191,
2831
+ "learning_rate": 2.0434782608695654e-07,
2832
+ "epoch": 4.214285714285714
2833
+ },
2834
+ {
2835
+ "current_steps": 472,
2836
+ "loss": 0.6678,
2837
+ "learning_rate": 2.0217391304347825e-07,
2838
+ "epoch": 4.223214285714286
2839
+ },
2840
+ {
2841
+ "current_steps": 473,
2842
+ "loss": 0.6296,
2843
+ "learning_rate": 2e-07,
2844
+ "epoch": 4.232142857142857
2845
+ },
2846
+ {
2847
+ "current_steps": 474,
2848
+ "loss": 0.884,
2849
+ "learning_rate": 1.9782608695652173e-07,
2850
+ "epoch": 4.241071428571429
2851
+ },
2852
+ {
2853
+ "current_steps": 475,
2854
+ "loss": 0.7207,
2855
+ "learning_rate": 1.9565217391304347e-07,
2856
+ "epoch": 4.25
2857
+ },
2858
+ {
2859
+ "current_steps": 476,
2860
+ "loss": 0.6856,
2861
+ "learning_rate": 1.934782608695652e-07,
2862
+ "epoch": 4.258928571428571
2863
+ },
2864
+ {
2865
+ "current_steps": 477,
2866
+ "loss": 0.6314,
2867
+ "learning_rate": 1.9130434782608695e-07,
2868
+ "epoch": 4.267857142857143
2869
+ },
2870
+ {
2871
+ "current_steps": 478,
2872
+ "loss": 0.5759,
2873
+ "learning_rate": 1.8913043478260867e-07,
2874
+ "epoch": 4.276785714285714
2875
+ },
2876
+ {
2877
+ "current_steps": 479,
2878
+ "loss": 0.6925,
2879
+ "learning_rate": 1.8695652173913043e-07,
2880
+ "epoch": 4.285714285714286
2881
+ },
2882
+ {
2883
+ "current_steps": 480,
2884
+ "loss": 0.6237,
2885
+ "learning_rate": 1.8478260869565215e-07,
2886
+ "epoch": 4.294642857142857
2887
+ },
2888
+ {
2889
+ "current_steps": 481,
2890
+ "loss": 0.6666,
2891
+ "learning_rate": 1.8260869565217391e-07,
2892
+ "epoch": 4.303571428571429
2893
+ },
2894
+ {
2895
+ "current_steps": 482,
2896
+ "loss": 0.709,
2897
+ "learning_rate": 1.8043478260869565e-07,
2898
+ "epoch": 4.3125
2899
+ },
2900
+ {
2901
+ "current_steps": 483,
2902
+ "loss": 0.8078,
2903
+ "learning_rate": 1.7826086956521737e-07,
2904
+ "epoch": 4.321428571428571
2905
+ },
2906
+ {
2907
+ "current_steps": 484,
2908
+ "loss": 0.7355,
2909
+ "learning_rate": 1.7608695652173914e-07,
2910
+ "epoch": 4.330357142857143
2911
+ },
2912
+ {
2913
+ "current_steps": 485,
2914
+ "loss": 0.8901,
2915
+ "learning_rate": 1.7391304347826085e-07,
2916
+ "epoch": 4.339285714285714
2917
+ },
2918
+ {
2919
+ "current_steps": 486,
2920
+ "loss": 0.565,
2921
+ "learning_rate": 1.7173913043478262e-07,
2922
+ "epoch": 4.348214285714286
2923
+ },
2924
+ {
2925
+ "current_steps": 487,
2926
+ "loss": 0.6396,
2927
+ "learning_rate": 1.6956521739130433e-07,
2928
+ "epoch": 4.357142857142857
2929
+ },
2930
+ {
2931
+ "current_steps": 488,
2932
+ "loss": 0.531,
2933
+ "learning_rate": 1.673913043478261e-07,
2934
+ "epoch": 4.366071428571429
2935
+ },
2936
+ {
2937
+ "current_steps": 489,
2938
+ "loss": 0.5726,
2939
+ "learning_rate": 1.652173913043478e-07,
2940
+ "epoch": 4.375
2941
+ },
2942
+ {
2943
+ "current_steps": 490,
2944
+ "loss": 0.602,
2945
+ "learning_rate": 1.6304347826086955e-07,
2946
+ "epoch": 4.383928571428571
2947
+ },
2948
+ {
2949
+ "current_steps": 491,
2950
+ "loss": 0.7032,
2951
+ "learning_rate": 1.608695652173913e-07,
2952
+ "epoch": 4.392857142857143
2953
+ },
2954
+ {
2955
+ "current_steps": 492,
2956
+ "loss": 0.8984,
2957
+ "learning_rate": 1.5869565217391303e-07,
2958
+ "epoch": 4.401785714285714
2959
+ },
2960
+ {
2961
+ "current_steps": 493,
2962
+ "loss": 0.5913,
2963
+ "learning_rate": 1.565217391304348e-07,
2964
+ "epoch": 4.410714285714286
2965
+ },
2966
+ {
2967
+ "current_steps": 494,
2968
+ "loss": 0.6021,
2969
+ "learning_rate": 1.543478260869565e-07,
2970
+ "epoch": 4.419642857142857
2971
+ },
2972
+ {
2973
+ "current_steps": 495,
2974
+ "loss": 0.7554,
2975
+ "learning_rate": 1.5217391304347825e-07,
2976
+ "epoch": 4.428571428571429
2977
+ },
2978
+ {
2979
+ "current_steps": 496,
2980
+ "loss": 0.8683,
2981
+ "learning_rate": 1.5e-07,
2982
+ "epoch": 4.4375
2983
+ },
2984
+ {
2985
+ "current_steps": 497,
2986
+ "loss": 0.5465,
2987
+ "learning_rate": 1.4782608695652173e-07,
2988
+ "epoch": 4.446428571428571
2989
+ },
2990
+ {
2991
+ "current_steps": 498,
2992
+ "loss": 0.6903,
2993
+ "learning_rate": 1.4565217391304347e-07,
2994
+ "epoch": 4.455357142857143
2995
+ },
2996
+ {
2997
+ "current_steps": 499,
2998
+ "loss": 0.4821,
2999
+ "learning_rate": 1.4347826086956521e-07,
3000
+ "epoch": 4.464285714285714
3001
+ },
3002
+ {
3003
+ "current_steps": 500,
3004
+ "loss": 0.6731,
3005
+ "learning_rate": 1.4130434782608693e-07,
3006
+ "epoch": 4.473214285714286
3007
+ },
3008
+ {
3009
+ "current_steps": 501,
3010
+ "loss": 0.7423,
3011
+ "learning_rate": 1.391304347826087e-07,
3012
+ "epoch": 4.482142857142857
3013
+ },
3014
+ {
3015
+ "current_steps": 502,
3016
+ "loss": 0.6967,
3017
+ "learning_rate": 1.369565217391304e-07,
3018
+ "epoch": 4.491071428571429
3019
+ },
3020
+ {
3021
+ "current_steps": 503,
3022
+ "loss": 0.5918,
3023
+ "learning_rate": 1.3478260869565218e-07,
3024
+ "epoch": 4.5
3025
+ },
3026
+ {
3027
+ "current_steps": 504,
3028
+ "loss": 0.8028,
3029
+ "learning_rate": 1.3260869565217392e-07,
3030
+ "epoch": 4.508928571428571
3031
+ },
3032
+ {
3033
+ "current_steps": 505,
3034
+ "loss": 0.9578,
3035
+ "learning_rate": 1.3043478260869563e-07,
3036
+ "epoch": 4.517857142857143
3037
+ },
3038
+ {
3039
+ "current_steps": 506,
3040
+ "loss": 0.6187,
3041
+ "learning_rate": 1.282608695652174e-07,
3042
+ "epoch": 4.526785714285714
3043
+ },
3044
+ {
3045
+ "current_steps": 507,
3046
+ "loss": 0.6426,
3047
+ "learning_rate": 1.260869565217391e-07,
3048
+ "epoch": 4.535714285714286
3049
+ },
3050
+ {
3051
+ "current_steps": 508,
3052
+ "loss": 0.5835,
3053
+ "learning_rate": 1.2391304347826088e-07,
3054
+ "epoch": 4.544642857142857
3055
+ },
3056
+ {
3057
+ "current_steps": 509,
3058
+ "loss": 0.7218,
3059
+ "learning_rate": 1.2173913043478262e-07,
3060
+ "epoch": 4.553571428571429
3061
+ },
3062
+ {
3063
+ "current_steps": 510,
3064
+ "loss": 0.812,
3065
+ "learning_rate": 1.1956521739130436e-07,
3066
+ "epoch": 4.5625
3067
+ },
3068
+ {
3069
+ "current_steps": 511,
3070
+ "loss": 0.5526,
3071
+ "learning_rate": 1.1739130434782609e-07,
3072
+ "epoch": 4.571428571428571
3073
+ },
3074
+ {
3075
+ "current_steps": 512,
3076
+ "loss": 0.8554,
3077
+ "learning_rate": 1.1521739130434783e-07,
3078
+ "epoch": 4.580357142857143
3079
+ },
3080
+ {
3081
+ "current_steps": 513,
3082
+ "loss": 0.7209,
3083
+ "learning_rate": 1.1304347826086955e-07,
3084
+ "epoch": 4.589285714285714
3085
+ },
3086
+ {
3087
+ "current_steps": 514,
3088
+ "loss": 0.7154,
3089
+ "learning_rate": 1.108695652173913e-07,
3090
+ "epoch": 4.598214285714286
3091
+ },
3092
+ {
3093
+ "current_steps": 515,
3094
+ "loss": 0.7147,
3095
+ "learning_rate": 1.0869565217391303e-07,
3096
+ "epoch": 4.607142857142857
3097
+ },
3098
+ {
3099
+ "current_steps": 516,
3100
+ "loss": 0.6997,
3101
+ "learning_rate": 1.0652173913043477e-07,
3102
+ "epoch": 4.616071428571429
3103
+ },
3104
+ {
3105
+ "current_steps": 517,
3106
+ "loss": 0.6283,
3107
+ "learning_rate": 1.0434782608695651e-07,
3108
+ "epoch": 4.625
3109
+ },
3110
+ {
3111
+ "current_steps": 518,
3112
+ "loss": 0.6279,
3113
+ "learning_rate": 1.0217391304347827e-07,
3114
+ "epoch": 4.633928571428571
3115
+ },
3116
+ {
3117
+ "current_steps": 519,
3118
+ "loss": 0.8152,
3119
+ "learning_rate": 1e-07,
3120
+ "epoch": 4.642857142857143
3121
+ },
3122
+ {
3123
+ "current_steps": 520,
3124
+ "loss": 0.6155,
3125
+ "learning_rate": 9.782608695652174e-08,
3126
+ "epoch": 4.651785714285714
3127
+ },
3128
+ {
3129
+ "current_steps": 521,
3130
+ "loss": 0.4727,
3131
+ "learning_rate": 9.565217391304348e-08,
3132
+ "epoch": 4.660714285714286
3133
+ },
3134
+ {
3135
+ "current_steps": 522,
3136
+ "loss": 0.7457,
3137
+ "learning_rate": 9.347826086956522e-08,
3138
+ "epoch": 4.669642857142857
3139
+ },
3140
+ {
3141
+ "current_steps": 523,
3142
+ "loss": 0.9712,
3143
+ "learning_rate": 9.130434782608696e-08,
3144
+ "epoch": 4.678571428571429
3145
+ },
3146
+ {
3147
+ "current_steps": 524,
3148
+ "loss": 0.7759,
3149
+ "learning_rate": 8.913043478260868e-08,
3150
+ "epoch": 4.6875
3151
+ },
3152
+ {
3153
+ "current_steps": 525,
3154
+ "loss": 0.6597,
3155
+ "learning_rate": 8.695652173913042e-08,
3156
+ "epoch": 4.696428571428571
3157
+ },
3158
+ {
3159
+ "current_steps": 526,
3160
+ "loss": 0.6258,
3161
+ "learning_rate": 8.478260869565216e-08,
3162
+ "epoch": 4.705357142857143
3163
+ },
3164
+ {
3165
+ "current_steps": 527,
3166
+ "loss": 0.6443,
3167
+ "learning_rate": 8.26086956521739e-08,
3168
+ "epoch": 4.714285714285714
3169
+ },
3170
+ {
3171
+ "current_steps": 528,
3172
+ "loss": 0.5547,
3173
+ "learning_rate": 8.043478260869565e-08,
3174
+ "epoch": 4.723214285714286
3175
+ },
3176
+ {
3177
+ "current_steps": 529,
3178
+ "loss": 0.7149,
3179
+ "learning_rate": 7.82608695652174e-08,
3180
+ "epoch": 4.732142857142857
3181
+ },
3182
+ {
3183
+ "current_steps": 530,
3184
+ "loss": 0.6138,
3185
+ "learning_rate": 7.608695652173913e-08,
3186
+ "epoch": 4.741071428571429
3187
+ },
3188
+ {
3189
+ "current_steps": 531,
3190
+ "loss": 0.8032,
3191
+ "learning_rate": 7.391304347826087e-08,
3192
+ "epoch": 4.75
3193
+ },
3194
+ {
3195
+ "current_steps": 532,
3196
+ "loss": 0.7141,
3197
+ "learning_rate": 7.173913043478261e-08,
3198
+ "epoch": 4.758928571428571
3199
+ },
3200
+ {
3201
+ "current_steps": 533,
3202
+ "loss": 0.724,
3203
+ "learning_rate": 6.956521739130435e-08,
3204
+ "epoch": 4.767857142857143
3205
+ },
3206
+ {
3207
+ "current_steps": 534,
3208
+ "loss": 0.7707,
3209
+ "learning_rate": 6.739130434782609e-08,
3210
+ "epoch": 4.776785714285714
3211
+ },
3212
+ {
3213
+ "current_steps": 535,
3214
+ "loss": 0.6754,
3215
+ "learning_rate": 6.521739130434782e-08,
3216
+ "epoch": 4.785714285714286
3217
+ },
3218
+ {
3219
+ "current_steps": 536,
3220
+ "loss": 0.5861,
3221
+ "learning_rate": 6.304347826086956e-08,
3222
+ "epoch": 4.794642857142857
3223
+ },
3224
+ {
3225
+ "current_steps": 537,
3226
+ "loss": 0.8395,
3227
+ "learning_rate": 6.086956521739131e-08,
3228
+ "epoch": 4.803571428571429
3229
+ },
3230
+ {
3231
+ "current_steps": 538,
3232
+ "loss": 0.7642,
3233
+ "learning_rate": 5.869565217391304e-08,
3234
+ "epoch": 4.8125
3235
+ },
3236
+ {
3237
+ "current_steps": 539,
3238
+ "loss": 0.735,
3239
+ "learning_rate": 5.6521739130434777e-08,
3240
+ "epoch": 4.821428571428571
3241
+ },
3242
+ {
3243
+ "current_steps": 540,
3244
+ "loss": 0.6153,
3245
+ "learning_rate": 5.434782608695652e-08,
3246
+ "epoch": 4.830357142857143
3247
+ },
3248
+ {
3249
+ "current_steps": 541,
3250
+ "loss": 0.6299,
3251
+ "learning_rate": 5.217391304347826e-08,
3252
+ "epoch": 4.839285714285714
3253
+ },
3254
+ {
3255
+ "current_steps": 542,
3256
+ "loss": 1.078,
3257
+ "learning_rate": 5e-08,
3258
+ "epoch": 4.848214285714286
3259
+ },
3260
+ {
3261
+ "current_steps": 543,
3262
+ "loss": 0.7314,
3263
+ "learning_rate": 4.782608695652174e-08,
3264
+ "epoch": 4.857142857142857
3265
+ },
3266
+ {
3267
+ "current_steps": 544,
3268
+ "loss": 0.8515,
3269
+ "learning_rate": 4.565217391304348e-08,
3270
+ "epoch": 4.866071428571429
3271
+ },
3272
+ {
3273
+ "current_steps": 545,
3274
+ "loss": 0.5401,
3275
+ "learning_rate": 4.347826086956521e-08,
3276
+ "epoch": 4.875
3277
+ },
3278
+ {
3279
+ "current_steps": 546,
3280
+ "loss": 0.7315,
3281
+ "learning_rate": 4.130434782608695e-08,
3282
+ "epoch": 4.883928571428571
3283
+ },
3284
+ {
3285
+ "current_steps": 547,
3286
+ "loss": 0.6113,
3287
+ "learning_rate": 3.91304347826087e-08,
3288
+ "epoch": 4.892857142857143
3289
+ },
3290
+ {
3291
+ "current_steps": 548,
3292
+ "loss": 0.6239,
3293
+ "learning_rate": 3.6956521739130433e-08,
3294
+ "epoch": 4.901785714285714
3295
+ },
3296
+ {
3297
+ "current_steps": 549,
3298
+ "loss": 0.7292,
3299
+ "learning_rate": 3.4782608695652174e-08,
3300
+ "epoch": 4.910714285714286
3301
+ },
3302
+ {
3303
+ "current_steps": 550,
3304
+ "loss": 0.5297,
3305
+ "learning_rate": 3.260869565217391e-08,
3306
+ "epoch": 4.919642857142857
3307
+ },
3308
+ {
3309
+ "current_steps": 551,
3310
+ "loss": 0.6269,
3311
+ "learning_rate": 3.0434782608695655e-08,
3312
+ "epoch": 4.928571428571429
3313
+ },
3314
+ {
3315
+ "current_steps": 552,
3316
+ "loss": 0.6724,
3317
+ "learning_rate": 2.8260869565217388e-08,
3318
+ "epoch": 4.9375
3319
+ },
3320
+ {
3321
+ "current_steps": 553,
3322
+ "loss": 0.5109,
3323
+ "learning_rate": 2.608695652173913e-08,
3324
+ "epoch": 4.946428571428571
3325
+ },
3326
+ {
3327
+ "current_steps": 554,
3328
+ "loss": 0.9446,
3329
+ "learning_rate": 2.391304347826087e-08,
3330
+ "epoch": 4.955357142857143
3331
+ },
3332
+ {
3333
+ "current_steps": 555,
3334
+ "loss": 0.6897,
3335
+ "learning_rate": 2.1739130434782606e-08,
3336
+ "epoch": 4.964285714285714
3337
+ },
3338
+ {
3339
+ "current_steps": 556,
3340
+ "loss": 0.5511,
3341
+ "learning_rate": 1.956521739130435e-08,
3342
+ "epoch": 4.973214285714286
3343
+ },
3344
+ {
3345
+ "current_steps": 557,
3346
+ "loss": 0.7246,
3347
+ "learning_rate": 1.7391304347826087e-08,
3348
+ "epoch": 4.982142857142857
3349
+ },
3350
+ {
3351
+ "current_steps": 558,
3352
+ "loss": 0.6332,
3353
+ "learning_rate": 1.5217391304347827e-08,
3354
+ "epoch": 4.991071428571429
3355
+ },
3356
+ {
3357
+ "current_steps": 559,
3358
+ "loss": 1.0499,
3359
+ "learning_rate": 1.3043478260869564e-08,
3360
+ "epoch": 5.0
3361
+ },
3362
+ {
3363
+ "current_steps": 559,
3364
+ "loss": 1.0499,
3365
+ "learning_rate": 1.3043478260869564e-08,
3366
+ "epoch": 5.0
3367
+ }
3368
+ ]
aliceinwonderland/training_graph.png ADDED
aliceinwonderland/training_log.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name": "Llama-2-13b-hf",
3
+ "base_model_class": "LlamaForCausalLM",
4
+ "base_loaded_in_4bit": true,
5
+ "base_loaded_in_8bit": false,
6
+ "projections": "q, v",
7
+ "loss": 1.0499,
8
+ "grad_norm": 5.645450592041016,
9
+ "learning_rate": 1.3043478260869564e-08,
10
+ "epoch": 5.0,
11
+ "current_steps": 559,
12
+ "current_steps_adjusted": 559,
13
+ "epoch_adjusted": 5.0,
14
+ "train_runtime": 1468.5439,
15
+ "train_samples_per_second": 1.515,
16
+ "train_steps_per_second": 0.381,
17
+ "total_flos": 4.4012668649472e+16,
18
+ "train_loss": 0.7355319578732763
19
+ }
aliceinwonderland/training_parameters.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "lora_name": "aliceinwonderland",
3
+ "always_override": true,
4
+ "save_steps": 0,
5
+ "micro_batch_size": 4,
6
+ "batch_size": 0,
7
+ "epochs": 5,
8
+ "learning_rate": "1e-6",
9
+ "lr_scheduler_type": "linear",
10
+ "lora_rank": 32,
11
+ "lora_alpha": 64,
12
+ "lora_dropout": 0.05,
13
+ "cutoff_len": 256,
14
+ "dataset": "None",
15
+ "eval_dataset": "None",
16
+ "format": "None",
17
+ "eval_steps": 100,
18
+ "raw_text_file": "aliceandwonderland",
19
+ "higher_rank_limit": false,
20
+ "warmup_steps": 100,
21
+ "optimizer": "adamw_torch",
22
+ "hard_cut_string": "\\n\\n\\n",
23
+ "train_only_after": "",
24
+ "stop_at_loss": 0,
25
+ "add_eos_token": false,
26
+ "min_chars": 20,
27
+ "report_to": "None",
28
+ "precize_slicing_overlap": true,
29
+ "add_eos_token_type": "Every Block",
30
+ "save_steps_under_loss": 1.8,
31
+ "add_bos_token": true,
32
+ "training_projection": "q-v",
33
+ "sliding_window": false,
34
+ "warmup_ratio": 0,
35
+ "grad_accumulation": 1,
36
+ "neft_noise_alpha": 0
37
+ }
aliceinwonderland/training_prompt.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "template_type": "raw_text"
3
+ }