Hugging Face's logo

ibru
/
bobo_groot_model

Safetensors
Model card Files
xet
 Community
ibru commited on
Commit
b43e31a
·
verified ·
1 Parent(s): 16e05ca

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-4000/README.md +202 -0
  2. checkpoint-4000/adapter_config.json +36 -0
  3. checkpoint-4000/adapter_model.safetensors +3 -0
  4. checkpoint-4000/experiment_cfg/metadata.json +275 -0
  5. checkpoint-4000/optimizer.pt +3 -0
  6. checkpoint-4000/rng_state.pth +3 -0
  7. checkpoint-4000/scheduler.pt +3 -0
  8. checkpoint-4000/trainer_state.json +2833 -0
  9. checkpoint-4500/README.md +202 -0
  10. checkpoint-4500/adapter_config.json +36 -0
  11. checkpoint-4500/adapter_model.safetensors +3 -0
  12. checkpoint-4500/experiment_cfg/metadata.json +275 -0
  13. checkpoint-4500/optimizer.pt +3 -0
  14. checkpoint-4500/rng_state.pth +3 -0
  15. checkpoint-4500/scheduler.pt +3 -0
  16. checkpoint-4500/trainer_state.json +3183 -0
  17. checkpoint-5000/README.md +202 -0
  18. checkpoint-5000/adapter_config.json +36 -0
  19. checkpoint-5000/adapter_model.safetensors +3 -0
  20. checkpoint-5000/experiment_cfg/metadata.json +275 -0
  21. checkpoint-5000/optimizer.pt +3 -0
  22. checkpoint-5000/rng_state.pth +3 -0
  23. checkpoint-5000/scheduler.pt +3 -0
  24. checkpoint-5000/trainer_state.json +3533 -0
  25. checkpoint-5500/README.md +202 -0
  26. checkpoint-5500/adapter_config.json +36 -0
  27. checkpoint-5500/adapter_model.safetensors +3 -0
  28. checkpoint-5500/experiment_cfg/metadata.json +275 -0
  29. checkpoint-5500/optimizer.pt +3 -0
  30. checkpoint-5500/rng_state.pth +3 -0
  31. checkpoint-5500/scheduler.pt +3 -0
  32. checkpoint-5500/trainer_state.json +3883 -0
  33. checkpoint-6000/README.md +202 -0
  34. checkpoint-6000/adapter_config.json +36 -0
  35. checkpoint-6000/adapter_model.safetensors +3 -0
  36. checkpoint-6000/experiment_cfg/metadata.json +275 -0
  37. checkpoint-6000/optimizer.pt +3 -0
  38. checkpoint-6000/rng_state.pth +3 -0
  39. checkpoint-6000/scheduler.pt +3 -0
  40. checkpoint-6000/trainer_state.json +0 -0
  41. checkpoint-6500/README.md +202 -0
  42. checkpoint-6500/adapter_config.json +36 -0
  43. checkpoint-6500/adapter_model.safetensors +3 -0
  44. checkpoint-6500/experiment_cfg/metadata.json +275 -0
  45. checkpoint-6500/optimizer.pt +3 -0
  46. checkpoint-6500/rng_state.pth +3 -0
  47. checkpoint-6500/scheduler.pt +3 -0
  48. checkpoint-6500/trainer_state.json +0 -0
  49. checkpoint-7000/README.md +202 -0
  50. checkpoint-7000/adapter_config.json +36 -0
checkpoint-4000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-4000/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-4000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf9900938297f10ca5bbfd3aea42b01ed8d414c257d79e23cf99c5abd7400bb4
3
+ size 123328576
checkpoint-4000/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-4000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8eeb6962209af3ac5e5780010e2e1e0f341c4fb0db66c4b4d9e43149a9fa7c5c
3
+ size 246824634
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff41b94b6c01cfd2a73909eff11b53105955ebf408577b11fc3fb83ad491ea1f
3
+ size 14244
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:217d233fe11f3792edfe1d9ed6553ae6ac505268a32c9fdc6ca353c28974200b
3
+ size 1064
checkpoint-4000/trainer_state.json ADDED
@@ -0,0 +1,2833 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 14.925373134328359,
5
+ "eval_steps": 500,
6
+ "global_step": 4000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03731343283582089,
13
+ "grad_norm": 0.8186072111129761,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 1.3847,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.07462686567164178,
20
+ "grad_norm": 0.5007426142692566,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 1.4283,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.11194029850746269,
27
+ "grad_norm": 0.49460887908935547,
28
+ "learning_rate": 3e-06,
29
+ "loss": 1.4868,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.14925373134328357,
34
+ "grad_norm": 0.5032920837402344,
35
+ "learning_rate": 4.000000000000001e-06,
36
+ "loss": 1.4491,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1865671641791045,
41
+ "grad_norm": 0.5688469409942627,
42
+ "learning_rate": 5e-06,
43
+ "loss": 1.3703,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.22388059701492538,
48
+ "grad_norm": 0.5052517652511597,
49
+ "learning_rate": 6e-06,
50
+ "loss": 1.419,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.26119402985074625,
55
+ "grad_norm": 0.6315643787384033,
56
+ "learning_rate": 7.000000000000001e-06,
57
+ "loss": 1.3058,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.29850746268656714,
62
+ "grad_norm": 0.6060447692871094,
63
+ "learning_rate": 8.000000000000001e-06,
64
+ "loss": 1.2908,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3358208955223881,
69
+ "grad_norm": 0.5513179302215576,
70
+ "learning_rate": 9e-06,
71
+ "loss": 1.2311,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.373134328358209,
76
+ "grad_norm": 0.8467404246330261,
77
+ "learning_rate": 1e-05,
78
+ "loss": 1.2043,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.41044776119402987,
83
+ "grad_norm": 0.8141824007034302,
84
+ "learning_rate": 1.1000000000000001e-05,
85
+ "loss": 1.0707,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.44776119402985076,
90
+ "grad_norm": 0.7932347059249878,
91
+ "learning_rate": 1.2e-05,
92
+ "loss": 0.9377,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.48507462686567165,
97
+ "grad_norm": 0.684220552444458,
98
+ "learning_rate": 1.3000000000000001e-05,
99
+ "loss": 0.714,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5223880597014925,
104
+ "grad_norm": 0.5886895060539246,
105
+ "learning_rate": 1.4000000000000001e-05,
106
+ "loss": 0.6479,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5597014925373134,
111
+ "grad_norm": 0.4764939248561859,
112
+ "learning_rate": 1.5e-05,
113
+ "loss": 0.5463,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.5970149253731343,
118
+ "grad_norm": 0.4621008038520813,
119
+ "learning_rate": 1.6000000000000003e-05,
120
+ "loss": 0.4641,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6343283582089553,
125
+ "grad_norm": 0.46492910385131836,
126
+ "learning_rate": 1.7000000000000003e-05,
127
+ "loss": 0.4159,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6716417910447762,
132
+ "grad_norm": 0.5017415881156921,
133
+ "learning_rate": 1.8e-05,
134
+ "loss": 0.4094,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7089552238805971,
139
+ "grad_norm": 0.34392210841178894,
140
+ "learning_rate": 1.9e-05,
141
+ "loss": 0.3478,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.746268656716418,
146
+ "grad_norm": 0.3240516483783722,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3821,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7835820895522388,
153
+ "grad_norm": 0.26301339268684387,
154
+ "learning_rate": 2.1e-05,
155
+ "loss": 0.3606,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8208955223880597,
160
+ "grad_norm": 0.34712520241737366,
161
+ "learning_rate": 2.2000000000000003e-05,
162
+ "loss": 0.3421,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8582089552238806,
167
+ "grad_norm": 0.3248469829559326,
168
+ "learning_rate": 2.3000000000000003e-05,
169
+ "loss": 0.3389,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.8955223880597015,
174
+ "grad_norm": 0.298149436712265,
175
+ "learning_rate": 2.4e-05,
176
+ "loss": 0.3145,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.9328358208955224,
181
+ "grad_norm": 0.2757190763950348,
182
+ "learning_rate": 2.5e-05,
183
+ "loss": 0.3065,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9701492537313433,
188
+ "grad_norm": 0.30510950088500977,
189
+ "learning_rate": 2.6000000000000002e-05,
190
+ "loss": 0.2971,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.007462686567164,
195
+ "grad_norm": 0.37349891662597656,
196
+ "learning_rate": 2.7000000000000002e-05,
197
+ "loss": 0.3273,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.044776119402985,
202
+ "grad_norm": 0.3667634129524231,
203
+ "learning_rate": 2.8000000000000003e-05,
204
+ "loss": 0.308,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.0820895522388059,
209
+ "grad_norm": 0.3463355004787445,
210
+ "learning_rate": 2.9e-05,
211
+ "loss": 0.3109,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.1194029850746268,
216
+ "grad_norm": 0.3888525366783142,
217
+ "learning_rate": 3e-05,
218
+ "loss": 0.2644,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.1567164179104479,
223
+ "grad_norm": 0.3749147951602936,
224
+ "learning_rate": 3.1e-05,
225
+ "loss": 0.2858,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.1940298507462686,
230
+ "grad_norm": 0.3270276188850403,
231
+ "learning_rate": 3.2000000000000005e-05,
232
+ "loss": 0.2573,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.2313432835820897,
237
+ "grad_norm": 0.3658592998981476,
238
+ "learning_rate": 3.3e-05,
239
+ "loss": 0.2613,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.2686567164179103,
244
+ "grad_norm": 0.3526328206062317,
245
+ "learning_rate": 3.4000000000000007e-05,
246
+ "loss": 0.2328,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.3059701492537314,
251
+ "grad_norm": 0.4528139531612396,
252
+ "learning_rate": 3.5e-05,
253
+ "loss": 0.2429,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.3432835820895521,
258
+ "grad_norm": 0.5426791310310364,
259
+ "learning_rate": 3.6e-05,
260
+ "loss": 0.2209,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.3805970149253732,
265
+ "grad_norm": 0.41844552755355835,
266
+ "learning_rate": 3.7e-05,
267
+ "loss": 0.2319,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.417910447761194,
272
+ "grad_norm": 0.4749431908130646,
273
+ "learning_rate": 3.8e-05,
274
+ "loss": 0.2233,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.455223880597015,
279
+ "grad_norm": 0.7010189890861511,
280
+ "learning_rate": 3.9000000000000006e-05,
281
+ "loss": 0.2181,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.4925373134328357,
286
+ "grad_norm": 0.5747635960578918,
287
+ "learning_rate": 4e-05,
288
+ "loss": 0.213,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.5298507462686568,
293
+ "grad_norm": 0.3661474287509918,
294
+ "learning_rate": 4.1e-05,
295
+ "loss": 0.2171,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.5671641791044775,
300
+ "grad_norm": 0.467835396528244,
301
+ "learning_rate": 4.2e-05,
302
+ "loss": 0.1985,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.6044776119402986,
307
+ "grad_norm": 0.5470123291015625,
308
+ "learning_rate": 4.3e-05,
309
+ "loss": 0.2176,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.6417910447761193,
314
+ "grad_norm": 0.5761199593544006,
315
+ "learning_rate": 4.4000000000000006e-05,
316
+ "loss": 0.2007,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.6791044776119404,
321
+ "grad_norm": 0.48257485032081604,
322
+ "learning_rate": 4.5e-05,
323
+ "loss": 0.2043,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.716417910447761,
328
+ "grad_norm": 0.48353052139282227,
329
+ "learning_rate": 4.600000000000001e-05,
330
+ "loss": 0.1872,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.7537313432835822,
335
+ "grad_norm": 0.4388391375541687,
336
+ "learning_rate": 4.7e-05,
337
+ "loss": 0.206,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.7910447761194028,
342
+ "grad_norm": 0.47332626581192017,
343
+ "learning_rate": 4.8e-05,
344
+ "loss": 0.1876,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.828358208955224,
349
+ "grad_norm": 0.8053535223007202,
350
+ "learning_rate": 4.9e-05,
351
+ "loss": 0.1839,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 1.8656716417910446,
356
+ "grad_norm": 0.413979709148407,
357
+ "learning_rate": 5e-05,
358
+ "loss": 0.1732,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 1.9029850746268657,
363
+ "grad_norm": 0.36910712718963623,
364
+ "learning_rate": 5.1000000000000006e-05,
365
+ "loss": 0.1827,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 1.9402985074626866,
370
+ "grad_norm": 0.8458298444747925,
371
+ "learning_rate": 5.2000000000000004e-05,
372
+ "loss": 0.1727,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 1.9776119402985075,
377
+ "grad_norm": 0.5452115535736084,
378
+ "learning_rate": 5.300000000000001e-05,
379
+ "loss": 0.1818,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.014925373134328,
384
+ "grad_norm": 0.4518108069896698,
385
+ "learning_rate": 5.4000000000000005e-05,
386
+ "loss": 0.177,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.0522388059701493,
391
+ "grad_norm": 0.66865074634552,
392
+ "learning_rate": 5.500000000000001e-05,
393
+ "loss": 0.1726,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.08955223880597,
398
+ "grad_norm": 0.6536034345626831,
399
+ "learning_rate": 5.6000000000000006e-05,
400
+ "loss": 0.1541,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.126865671641791,
405
+ "grad_norm": 0.5571377277374268,
406
+ "learning_rate": 5.6999999999999996e-05,
407
+ "loss": 0.1671,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.1641791044776117,
412
+ "grad_norm": 0.5385546684265137,
413
+ "learning_rate": 5.8e-05,
414
+ "loss": 0.1582,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.201492537313433,
419
+ "grad_norm": 0.577961266040802,
420
+ "learning_rate": 5.9e-05,
421
+ "loss": 0.1528,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.2388059701492535,
426
+ "grad_norm": 0.5082416534423828,
427
+ "learning_rate": 6e-05,
428
+ "loss": 0.1638,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.2761194029850746,
433
+ "grad_norm": 0.5490861535072327,
434
+ "learning_rate": 6.1e-05,
435
+ "loss": 0.166,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.3134328358208958,
440
+ "grad_norm": 0.492366760969162,
441
+ "learning_rate": 6.2e-05,
442
+ "loss": 0.1481,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.3507462686567164,
447
+ "grad_norm": 0.3702855110168457,
448
+ "learning_rate": 6.3e-05,
449
+ "loss": 0.1514,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.388059701492537,
454
+ "grad_norm": 0.664667010307312,
455
+ "learning_rate": 6.400000000000001e-05,
456
+ "loss": 0.1441,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.425373134328358,
461
+ "grad_norm": 0.33382174372673035,
462
+ "learning_rate": 6.500000000000001e-05,
463
+ "loss": 0.1573,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.4626865671641793,
468
+ "grad_norm": 0.4848814010620117,
469
+ "learning_rate": 6.6e-05,
470
+ "loss": 0.1457,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.5,
475
+ "grad_norm": 0.3649997413158417,
476
+ "learning_rate": 6.7e-05,
477
+ "loss": 0.1467,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.5373134328358207,
482
+ "grad_norm": 0.6385223865509033,
483
+ "learning_rate": 6.800000000000001e-05,
484
+ "loss": 0.145,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.574626865671642,
489
+ "grad_norm": 0.4580625891685486,
490
+ "learning_rate": 6.9e-05,
491
+ "loss": 0.1352,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.611940298507463,
496
+ "grad_norm": 0.5141746401786804,
497
+ "learning_rate": 7e-05,
498
+ "loss": 0.1444,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.6492537313432836,
503
+ "grad_norm": 0.40220722556114197,
504
+ "learning_rate": 7.1e-05,
505
+ "loss": 0.1493,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.6865671641791042,
510
+ "grad_norm": 0.5510571002960205,
511
+ "learning_rate": 7.2e-05,
512
+ "loss": 0.1387,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.7238805970149254,
517
+ "grad_norm": 0.43814659118652344,
518
+ "learning_rate": 7.3e-05,
519
+ "loss": 0.1374,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 2.7611940298507465,
524
+ "grad_norm": 0.4118008613586426,
525
+ "learning_rate": 7.4e-05,
526
+ "loss": 0.1297,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 2.798507462686567,
531
+ "grad_norm": 0.5626503229141235,
532
+ "learning_rate": 7.500000000000001e-05,
533
+ "loss": 0.1299,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 2.835820895522388,
538
+ "grad_norm": 0.4066360592842102,
539
+ "learning_rate": 7.6e-05,
540
+ "loss": 0.1102,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 2.873134328358209,
545
+ "grad_norm": 0.47184985876083374,
546
+ "learning_rate": 7.7e-05,
547
+ "loss": 0.1219,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 2.91044776119403,
552
+ "grad_norm": 0.6611475348472595,
553
+ "learning_rate": 7.800000000000001e-05,
554
+ "loss": 0.1267,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 2.9477611940298507,
559
+ "grad_norm": 0.3570108413696289,
560
+ "learning_rate": 7.900000000000001e-05,
561
+ "loss": 0.1191,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 2.9850746268656714,
566
+ "grad_norm": 0.4581681489944458,
567
+ "learning_rate": 8e-05,
568
+ "loss": 0.1209,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.0223880597014925,
573
+ "grad_norm": 0.4643435776233673,
574
+ "learning_rate": 8.1e-05,
575
+ "loss": 0.129,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.0597014925373136,
580
+ "grad_norm": 0.5595763921737671,
581
+ "learning_rate": 8.2e-05,
582
+ "loss": 0.1158,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.0970149253731343,
587
+ "grad_norm": 0.48848605155944824,
588
+ "learning_rate": 8.3e-05,
589
+ "loss": 0.1188,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.1343283582089554,
594
+ "grad_norm": 0.4496570825576782,
595
+ "learning_rate": 8.4e-05,
596
+ "loss": 0.114,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.171641791044776,
601
+ "grad_norm": 0.31364986300468445,
602
+ "learning_rate": 8.5e-05,
603
+ "loss": 0.1196,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.208955223880597,
608
+ "grad_norm": 0.3395878076553345,
609
+ "learning_rate": 8.6e-05,
610
+ "loss": 0.1124,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.246268656716418,
615
+ "grad_norm": 0.4917413592338562,
616
+ "learning_rate": 8.7e-05,
617
+ "loss": 0.1074,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.283582089552239,
622
+ "grad_norm": 0.44114553928375244,
623
+ "learning_rate": 8.800000000000001e-05,
624
+ "loss": 0.1095,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.3208955223880596,
629
+ "grad_norm": 0.3323831558227539,
630
+ "learning_rate": 8.900000000000001e-05,
631
+ "loss": 0.106,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.3582089552238807,
636
+ "grad_norm": 0.4495660066604614,
637
+ "learning_rate": 9e-05,
638
+ "loss": 0.1222,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.3955223880597014,
643
+ "grad_norm": 0.40784788131713867,
644
+ "learning_rate": 9.1e-05,
645
+ "loss": 0.1048,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.4328358208955225,
650
+ "grad_norm": 0.4643700420856476,
651
+ "learning_rate": 9.200000000000001e-05,
652
+ "loss": 0.1097,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.470149253731343,
657
+ "grad_norm": 0.472494512796402,
658
+ "learning_rate": 9.300000000000001e-05,
659
+ "loss": 0.1041,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.5074626865671643,
664
+ "grad_norm": 0.6110897660255432,
665
+ "learning_rate": 9.4e-05,
666
+ "loss": 0.0959,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.544776119402985,
671
+ "grad_norm": 0.5313069820404053,
672
+ "learning_rate": 9.5e-05,
673
+ "loss": 0.113,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.582089552238806,
678
+ "grad_norm": 0.4223133623600006,
679
+ "learning_rate": 9.6e-05,
680
+ "loss": 0.099,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.6194029850746268,
685
+ "grad_norm": 0.5464731454849243,
686
+ "learning_rate": 9.7e-05,
687
+ "loss": 0.1008,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 3.656716417910448,
692
+ "grad_norm": 0.3538314402103424,
693
+ "learning_rate": 9.8e-05,
694
+ "loss": 0.1049,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 3.6940298507462686,
699
+ "grad_norm": 0.7460148334503174,
700
+ "learning_rate": 9.900000000000001e-05,
701
+ "loss": 0.1088,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 3.7313432835820897,
706
+ "grad_norm": 0.3210597038269043,
707
+ "learning_rate": 0.0001,
708
+ "loss": 0.1041,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 3.7686567164179103,
713
+ "grad_norm": 0.4450497627258301,
714
+ "learning_rate": 9.999993165095463e-05,
715
+ "loss": 0.0985,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 3.8059701492537314,
720
+ "grad_norm": 0.4348960816860199,
721
+ "learning_rate": 9.999972660400536e-05,
722
+ "loss": 0.1015,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 3.843283582089552,
727
+ "grad_norm": 0.462782621383667,
728
+ "learning_rate": 9.999938485971279e-05,
729
+ "loss": 0.1068,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 3.8805970149253732,
734
+ "grad_norm": 0.3801368474960327,
735
+ "learning_rate": 9.999890641901125e-05,
736
+ "loss": 0.1117,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 3.917910447761194,
741
+ "grad_norm": 0.45135366916656494,
742
+ "learning_rate": 9.999829128320874e-05,
743
+ "loss": 0.0917,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 3.955223880597015,
748
+ "grad_norm": 0.41138389706611633,
749
+ "learning_rate": 9.999753945398704e-05,
750
+ "loss": 0.1049,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 3.9925373134328357,
755
+ "grad_norm": 0.4976252317428589,
756
+ "learning_rate": 9.999665093340165e-05,
757
+ "loss": 0.1029,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.029850746268656,
762
+ "grad_norm": 0.46372008323669434,
763
+ "learning_rate": 9.99956257238817e-05,
764
+ "loss": 0.1012,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.067164179104478,
769
+ "grad_norm": 0.546938955783844,
770
+ "learning_rate": 9.999446382823013e-05,
771
+ "loss": 0.0829,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.104477611940299,
776
+ "grad_norm": 0.40513405203819275,
777
+ "learning_rate": 9.999316524962345e-05,
778
+ "loss": 0.0933,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.141791044776119,
783
+ "grad_norm": 0.4198484420776367,
784
+ "learning_rate": 9.999172999161198e-05,
785
+ "loss": 0.0895,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.17910447761194,
790
+ "grad_norm": 0.3965628743171692,
791
+ "learning_rate": 9.999015805811965e-05,
792
+ "loss": 0.0917,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.2164179104477615,
797
+ "grad_norm": 0.3095884621143341,
798
+ "learning_rate": 9.998844945344405e-05,
799
+ "loss": 0.0953,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.253731343283582,
804
+ "grad_norm": 0.7962276339530945,
805
+ "learning_rate": 9.998660418225645e-05,
806
+ "loss": 0.0979,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.291044776119403,
811
+ "grad_norm": 0.42066490650177,
812
+ "learning_rate": 9.998462224960175e-05,
813
+ "loss": 0.099,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.3283582089552235,
818
+ "grad_norm": 0.3894193470478058,
819
+ "learning_rate": 9.998250366089848e-05,
820
+ "loss": 0.0887,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.365671641791045,
825
+ "grad_norm": 0.28998032212257385,
826
+ "learning_rate": 9.998024842193876e-05,
827
+ "loss": 0.0943,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.402985074626866,
832
+ "grad_norm": 0.3919823467731476,
833
+ "learning_rate": 9.997785653888835e-05,
834
+ "loss": 0.0916,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.440298507462686,
839
+ "grad_norm": 0.3708650469779968,
840
+ "learning_rate": 9.997532801828658e-05,
841
+ "loss": 0.0858,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.477611940298507,
846
+ "grad_norm": 0.2935069799423218,
847
+ "learning_rate": 9.997266286704631e-05,
848
+ "loss": 0.0992,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.514925373134329,
853
+ "grad_norm": 0.4675377607345581,
854
+ "learning_rate": 9.996986109245395e-05,
855
+ "loss": 0.0854,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.552238805970149,
860
+ "grad_norm": 0.31374865770339966,
861
+ "learning_rate": 9.996692270216947e-05,
862
+ "loss": 0.0788,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 4.58955223880597,
867
+ "grad_norm": 0.419249951839447,
868
+ "learning_rate": 9.996384770422629e-05,
869
+ "loss": 0.0873,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 4.6268656716417915,
874
+ "grad_norm": 0.26002731919288635,
875
+ "learning_rate": 9.996063610703137e-05,
876
+ "loss": 0.0845,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 4.664179104477612,
881
+ "grad_norm": 0.29573896527290344,
882
+ "learning_rate": 9.995728791936504e-05,
883
+ "loss": 0.091,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 4.701492537313433,
888
+ "grad_norm": 0.33090147376060486,
889
+ "learning_rate": 9.995380315038119e-05,
890
+ "loss": 0.0827,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 4.7388059701492535,
895
+ "grad_norm": 0.24417485296726227,
896
+ "learning_rate": 9.9950181809607e-05,
897
+ "loss": 0.0859,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 4.776119402985074,
902
+ "grad_norm": 0.48290401697158813,
903
+ "learning_rate": 9.994642390694308e-05,
904
+ "loss": 0.0889,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 4.813432835820896,
909
+ "grad_norm": 0.4479697048664093,
910
+ "learning_rate": 9.99425294526634e-05,
911
+ "loss": 0.097,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 4.850746268656716,
916
+ "grad_norm": 0.3560147285461426,
917
+ "learning_rate": 9.993849845741524e-05,
918
+ "loss": 0.0904,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 4.888059701492537,
923
+ "grad_norm": 0.6645416617393494,
924
+ "learning_rate": 9.99343309322192e-05,
925
+ "loss": 0.0922,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 4.925373134328359,
930
+ "grad_norm": 0.29696759581565857,
931
+ "learning_rate": 9.993002688846913e-05,
932
+ "loss": 0.093,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 4.962686567164179,
937
+ "grad_norm": 0.47146692872047424,
938
+ "learning_rate": 9.992558633793212e-05,
939
+ "loss": 0.085,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.0,
944
+ "grad_norm": 0.3430916368961334,
945
+ "learning_rate": 9.992100929274846e-05,
946
+ "loss": 0.0805,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.037313432835821,
951
+ "grad_norm": 0.3205055892467499,
952
+ "learning_rate": 9.991629576543163e-05,
953
+ "loss": 0.0766,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.074626865671641,
958
+ "grad_norm": 0.3664805293083191,
959
+ "learning_rate": 9.991144576886823e-05,
960
+ "loss": 0.0766,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.111940298507463,
965
+ "grad_norm": 0.3753412663936615,
966
+ "learning_rate": 9.990645931631796e-05,
967
+ "loss": 0.0688,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.149253731343284,
972
+ "grad_norm": 0.31633055210113525,
973
+ "learning_rate": 9.990133642141359e-05,
974
+ "loss": 0.0796,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.186567164179104,
979
+ "grad_norm": 0.3355732262134552,
980
+ "learning_rate": 9.989607709816091e-05,
981
+ "loss": 0.0716,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.223880597014926,
986
+ "grad_norm": 0.24850831925868988,
987
+ "learning_rate": 9.989068136093873e-05,
988
+ "loss": 0.0778,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.2611940298507465,
993
+ "grad_norm": 0.29537102580070496,
994
+ "learning_rate": 9.988514922449879e-05,
995
+ "loss": 0.0759,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.298507462686567,
1000
+ "grad_norm": 0.3430945873260498,
1001
+ "learning_rate": 9.987948070396571e-05,
1002
+ "loss": 0.0774,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.335820895522388,
1007
+ "grad_norm": 0.5220637917518616,
1008
+ "learning_rate": 9.987367581483705e-05,
1009
+ "loss": 0.0836,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.373134328358209,
1014
+ "grad_norm": 0.28184008598327637,
1015
+ "learning_rate": 9.986773457298311e-05,
1016
+ "loss": 0.0752,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.41044776119403,
1021
+ "grad_norm": 0.36261311173439026,
1022
+ "learning_rate": 9.986165699464705e-05,
1023
+ "loss": 0.075,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.447761194029851,
1028
+ "grad_norm": 0.5107380151748657,
1029
+ "learning_rate": 9.985544309644475e-05,
1030
+ "loss": 0.0814,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 5.485074626865671,
1035
+ "grad_norm": 0.2446671426296234,
1036
+ "learning_rate": 9.984909289536473e-05,
1037
+ "loss": 0.0704,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 5.522388059701493,
1042
+ "grad_norm": 0.30449381470680237,
1043
+ "learning_rate": 9.984260640876821e-05,
1044
+ "loss": 0.0794,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 5.559701492537314,
1049
+ "grad_norm": 0.25645050406455994,
1050
+ "learning_rate": 9.983598365438902e-05,
1051
+ "loss": 0.0709,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 5.597014925373134,
1056
+ "grad_norm": 0.23825006186962128,
1057
+ "learning_rate": 9.98292246503335e-05,
1058
+ "loss": 0.0828,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 5.634328358208955,
1063
+ "grad_norm": 0.3259269893169403,
1064
+ "learning_rate": 9.98223294150805e-05,
1065
+ "loss": 0.0824,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 5.6716417910447765,
1070
+ "grad_norm": 0.24058914184570312,
1071
+ "learning_rate": 9.981529796748134e-05,
1072
+ "loss": 0.073,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 5.708955223880597,
1077
+ "grad_norm": 0.34457242488861084,
1078
+ "learning_rate": 9.980813032675974e-05,
1079
+ "loss": 0.0845,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 5.746268656716418,
1084
+ "grad_norm": 0.32940393686294556,
1085
+ "learning_rate": 9.980082651251175e-05,
1086
+ "loss": 0.0832,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 5.7835820895522385,
1091
+ "grad_norm": 0.5683007836341858,
1092
+ "learning_rate": 9.979338654470569e-05,
1093
+ "loss": 0.0836,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 5.82089552238806,
1098
+ "grad_norm": 0.31041061878204346,
1099
+ "learning_rate": 9.97858104436822e-05,
1100
+ "loss": 0.07,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 5.858208955223881,
1105
+ "grad_norm": 0.37858131527900696,
1106
+ "learning_rate": 9.977809823015401e-05,
1107
+ "loss": 0.0738,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 5.895522388059701,
1112
+ "grad_norm": 0.2743091583251953,
1113
+ "learning_rate": 9.977024992520602e-05,
1114
+ "loss": 0.0761,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 5.932835820895522,
1119
+ "grad_norm": 0.29117098450660706,
1120
+ "learning_rate": 9.976226555029522e-05,
1121
+ "loss": 0.0777,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 5.970149253731344,
1126
+ "grad_norm": 0.31398633122444153,
1127
+ "learning_rate": 9.975414512725057e-05,
1128
+ "loss": 0.0664,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.007462686567164,
1133
+ "grad_norm": 0.2684272527694702,
1134
+ "learning_rate": 9.974588867827301e-05,
1135
+ "loss": 0.0686,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.044776119402985,
1140
+ "grad_norm": 0.3945397436618805,
1141
+ "learning_rate": 9.973749622593534e-05,
1142
+ "loss": 0.0614,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.082089552238806,
1147
+ "grad_norm": 0.2747954726219177,
1148
+ "learning_rate": 9.972896779318219e-05,
1149
+ "loss": 0.0681,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.119402985074627,
1154
+ "grad_norm": 0.43257200717926025,
1155
+ "learning_rate": 9.972030340333001e-05,
1156
+ "loss": 0.0725,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.156716417910448,
1161
+ "grad_norm": 0.3559250831604004,
1162
+ "learning_rate": 9.97115030800669e-05,
1163
+ "loss": 0.0804,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.1940298507462686,
1168
+ "grad_norm": 0.3079264760017395,
1169
+ "learning_rate": 9.970256684745258e-05,
1170
+ "loss": 0.0649,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.231343283582089,
1175
+ "grad_norm": 0.32298946380615234,
1176
+ "learning_rate": 9.969349472991838e-05,
1177
+ "loss": 0.0668,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.268656716417911,
1182
+ "grad_norm": 0.2826225459575653,
1183
+ "learning_rate": 9.968428675226714e-05,
1184
+ "loss": 0.0734,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.3059701492537314,
1189
+ "grad_norm": 0.39002349972724915,
1190
+ "learning_rate": 9.967494293967312e-05,
1191
+ "loss": 0.0728,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.343283582089552,
1196
+ "grad_norm": 0.403890997171402,
1197
+ "learning_rate": 9.966546331768191e-05,
1198
+ "loss": 0.067,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.380597014925373,
1203
+ "grad_norm": 0.3755359351634979,
1204
+ "learning_rate": 9.965584791221048e-05,
1205
+ "loss": 0.0755,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 6.417910447761194,
1210
+ "grad_norm": 0.26346635818481445,
1211
+ "learning_rate": 9.964609674954696e-05,
1212
+ "loss": 0.0728,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 6.455223880597015,
1217
+ "grad_norm": 0.45292145013809204,
1218
+ "learning_rate": 9.963620985635065e-05,
1219
+ "loss": 0.0731,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 6.492537313432836,
1224
+ "grad_norm": 0.3568434715270996,
1225
+ "learning_rate": 9.962618725965196e-05,
1226
+ "loss": 0.0761,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 6.529850746268656,
1231
+ "grad_norm": 0.2551257014274597,
1232
+ "learning_rate": 9.961602898685226e-05,
1233
+ "loss": 0.0694,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 6.567164179104478,
1238
+ "grad_norm": 0.6106354594230652,
1239
+ "learning_rate": 9.96057350657239e-05,
1240
+ "loss": 0.0827,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 6.604477611940299,
1245
+ "grad_norm": 0.3226093053817749,
1246
+ "learning_rate": 9.959530552441005e-05,
1247
+ "loss": 0.0716,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 6.641791044776119,
1252
+ "grad_norm": 0.4297254979610443,
1253
+ "learning_rate": 9.95847403914247e-05,
1254
+ "loss": 0.0748,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 6.67910447761194,
1259
+ "grad_norm": 0.26469680666923523,
1260
+ "learning_rate": 9.95740396956525e-05,
1261
+ "loss": 0.074,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 6.7164179104477615,
1266
+ "grad_norm": 0.22717897593975067,
1267
+ "learning_rate": 9.956320346634876e-05,
1268
+ "loss": 0.0739,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 6.753731343283582,
1273
+ "grad_norm": 0.4513498544692993,
1274
+ "learning_rate": 9.955223173313931e-05,
1275
+ "loss": 0.0664,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 6.791044776119403,
1280
+ "grad_norm": 0.31683439016342163,
1281
+ "learning_rate": 9.954112452602045e-05,
1282
+ "loss": 0.069,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 6.8283582089552235,
1287
+ "grad_norm": 0.3350532650947571,
1288
+ "learning_rate": 9.952988187535886e-05,
1289
+ "loss": 0.0699,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 6.865671641791045,
1294
+ "grad_norm": 0.29829463362693787,
1295
+ "learning_rate": 9.95185038118915e-05,
1296
+ "loss": 0.0663,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 6.902985074626866,
1301
+ "grad_norm": 0.31650781631469727,
1302
+ "learning_rate": 9.950699036672559e-05,
1303
+ "loss": 0.0668,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 6.940298507462686,
1308
+ "grad_norm": 0.360944926738739,
1309
+ "learning_rate": 9.949534157133844e-05,
1310
+ "loss": 0.0696,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 6.977611940298507,
1315
+ "grad_norm": 0.31337013840675354,
1316
+ "learning_rate": 9.948355745757741e-05,
1317
+ "loss": 0.073,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.014925373134329,
1322
+ "grad_norm": 0.4675919711589813,
1323
+ "learning_rate": 9.94716380576598e-05,
1324
+ "loss": 0.0688,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.052238805970149,
1329
+ "grad_norm": 0.3031919002532959,
1330
+ "learning_rate": 9.945958340417283e-05,
1331
+ "loss": 0.0596,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.08955223880597,
1336
+ "grad_norm": 0.24858474731445312,
1337
+ "learning_rate": 9.944739353007344e-05,
1338
+ "loss": 0.0717,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.126865671641791,
1343
+ "grad_norm": 0.20959483087062836,
1344
+ "learning_rate": 9.943506846868826e-05,
1345
+ "loss": 0.0694,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.164179104477612,
1350
+ "grad_norm": 0.35621434450149536,
1351
+ "learning_rate": 9.942260825371358e-05,
1352
+ "loss": 0.063,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.201492537313433,
1357
+ "grad_norm": 0.3462587594985962,
1358
+ "learning_rate": 9.941001291921512e-05,
1359
+ "loss": 0.068,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.2388059701492535,
1364
+ "grad_norm": 0.38649681210517883,
1365
+ "learning_rate": 9.939728249962807e-05,
1366
+ "loss": 0.0638,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.276119402985074,
1371
+ "grad_norm": 0.29564595222473145,
1372
+ "learning_rate": 9.938441702975689e-05,
1373
+ "loss": 0.0626,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 7.313432835820896,
1378
+ "grad_norm": 0.339857816696167,
1379
+ "learning_rate": 9.937141654477528e-05,
1380
+ "loss": 0.0535,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 7.350746268656716,
1385
+ "grad_norm": 0.2591215670108795,
1386
+ "learning_rate": 9.93582810802261e-05,
1387
+ "loss": 0.0645,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 7.388059701492537,
1392
+ "grad_norm": 0.30237796902656555,
1393
+ "learning_rate": 9.934501067202117e-05,
1394
+ "loss": 0.0675,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 7.425373134328359,
1399
+ "grad_norm": 0.28394174575805664,
1400
+ "learning_rate": 9.93316053564413e-05,
1401
+ "loss": 0.0643,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 7.462686567164179,
1406
+ "grad_norm": 0.3124663233757019,
1407
+ "learning_rate": 9.931806517013612e-05,
1408
+ "loss": 0.059,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 7.5,
1413
+ "grad_norm": 0.36073037981987,
1414
+ "learning_rate": 9.930439015012396e-05,
1415
+ "loss": 0.0606,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 7.537313432835821,
1420
+ "grad_norm": 0.4091481864452362,
1421
+ "learning_rate": 9.929058033379181e-05,
1422
+ "loss": 0.0603,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 7.574626865671641,
1427
+ "grad_norm": 0.44718074798583984,
1428
+ "learning_rate": 9.927663575889521e-05,
1429
+ "loss": 0.0741,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 7.611940298507463,
1434
+ "grad_norm": 0.3819601833820343,
1435
+ "learning_rate": 9.926255646355804e-05,
1436
+ "loss": 0.0707,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 7.649253731343284,
1441
+ "grad_norm": 0.23336420953273773,
1442
+ "learning_rate": 9.92483424862726e-05,
1443
+ "loss": 0.0676,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 7.686567164179104,
1448
+ "grad_norm": 0.24415315687656403,
1449
+ "learning_rate": 9.923399386589933e-05,
1450
+ "loss": 0.0594,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 7.723880597014926,
1455
+ "grad_norm": 0.3735473155975342,
1456
+ "learning_rate": 9.921951064166684e-05,
1457
+ "loss": 0.062,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 7.7611940298507465,
1462
+ "grad_norm": 0.31629472970962524,
1463
+ "learning_rate": 9.92048928531717e-05,
1464
+ "loss": 0.0606,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 7.798507462686567,
1469
+ "grad_norm": 0.37902557849884033,
1470
+ "learning_rate": 9.919014054037836e-05,
1471
+ "loss": 0.0584,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 7.835820895522388,
1476
+ "grad_norm": 0.3486720323562622,
1477
+ "learning_rate": 9.917525374361912e-05,
1478
+ "loss": 0.056,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 7.8731343283582085,
1483
+ "grad_norm": 0.3731362521648407,
1484
+ "learning_rate": 9.91602325035939e-05,
1485
+ "loss": 0.0601,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 7.91044776119403,
1490
+ "grad_norm": 0.3560399115085602,
1491
+ "learning_rate": 9.914507686137019e-05,
1492
+ "loss": 0.06,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 7.947761194029851,
1497
+ "grad_norm": 0.30075564980506897,
1498
+ "learning_rate": 9.912978685838294e-05,
1499
+ "loss": 0.0657,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 7.985074626865671,
1504
+ "grad_norm": 0.2984028458595276,
1505
+ "learning_rate": 9.911436253643445e-05,
1506
+ "loss": 0.0587,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 8.022388059701493,
1511
+ "grad_norm": 0.1980169117450714,
1512
+ "learning_rate": 9.90988039376942e-05,
1513
+ "loss": 0.0718,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 8.059701492537313,
1518
+ "grad_norm": 0.31339579820632935,
1519
+ "learning_rate": 9.90831111046988e-05,
1520
+ "loss": 0.0557,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 8.097014925373134,
1525
+ "grad_norm": 0.1968696266412735,
1526
+ "learning_rate": 9.90672840803519e-05,
1527
+ "loss": 0.0571,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 8.134328358208956,
1532
+ "grad_norm": 0.23931682109832764,
1533
+ "learning_rate": 9.905132290792394e-05,
1534
+ "loss": 0.0566,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 8.171641791044776,
1539
+ "grad_norm": 0.21741189062595367,
1540
+ "learning_rate": 9.903522763105218e-05,
1541
+ "loss": 0.0575,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 8.208955223880597,
1546
+ "grad_norm": 0.22874368727207184,
1547
+ "learning_rate": 9.901899829374047e-05,
1548
+ "loss": 0.0565,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 8.246268656716419,
1553
+ "grad_norm": 0.3441888093948364,
1554
+ "learning_rate": 9.900263494035921e-05,
1555
+ "loss": 0.0565,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 8.283582089552239,
1560
+ "grad_norm": 0.2539830803871155,
1561
+ "learning_rate": 9.89861376156452e-05,
1562
+ "loss": 0.0538,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 8.32089552238806,
1567
+ "grad_norm": 0.2235102653503418,
1568
+ "learning_rate": 9.896950636470147e-05,
1569
+ "loss": 0.0609,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 8.35820895522388,
1574
+ "grad_norm": 0.1941322684288025,
1575
+ "learning_rate": 9.895274123299723e-05,
1576
+ "loss": 0.0562,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 8.395522388059701,
1581
+ "grad_norm": 0.2691369950771332,
1582
+ "learning_rate": 9.893584226636772e-05,
1583
+ "loss": 0.0608,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 8.432835820895523,
1588
+ "grad_norm": 0.24730461835861206,
1589
+ "learning_rate": 9.891880951101407e-05,
1590
+ "loss": 0.0582,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 8.470149253731343,
1595
+ "grad_norm": 0.34785839915275574,
1596
+ "learning_rate": 9.890164301350318e-05,
1597
+ "loss": 0.0506,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 8.507462686567164,
1602
+ "grad_norm": 0.3625825345516205,
1603
+ "learning_rate": 9.888434282076758e-05,
1604
+ "loss": 0.0614,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 8.544776119402986,
1609
+ "grad_norm": 0.25210148096084595,
1610
+ "learning_rate": 9.886690898010535e-05,
1611
+ "loss": 0.0611,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 8.582089552238806,
1616
+ "grad_norm": 0.27312466502189636,
1617
+ "learning_rate": 9.884934153917997e-05,
1618
+ "loss": 0.0537,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 8.619402985074627,
1623
+ "grad_norm": 0.314647912979126,
1624
+ "learning_rate": 9.883164054602012e-05,
1625
+ "loss": 0.0602,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 8.656716417910447,
1630
+ "grad_norm": 0.21531912684440613,
1631
+ "learning_rate": 9.881380604901964e-05,
1632
+ "loss": 0.0552,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 8.694029850746269,
1637
+ "grad_norm": 0.23920664191246033,
1638
+ "learning_rate": 9.879583809693738e-05,
1639
+ "loss": 0.0613,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 8.73134328358209,
1644
+ "grad_norm": 0.21864956617355347,
1645
+ "learning_rate": 9.877773673889701e-05,
1646
+ "loss": 0.0649,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 8.76865671641791,
1651
+ "grad_norm": 0.27523377537727356,
1652
+ "learning_rate": 9.8759502024387e-05,
1653
+ "loss": 0.0606,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 8.805970149253731,
1658
+ "grad_norm": 0.24805469810962677,
1659
+ "learning_rate": 9.87411340032603e-05,
1660
+ "loss": 0.0549,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 8.843283582089553,
1665
+ "grad_norm": 0.23070092499256134,
1666
+ "learning_rate": 9.872263272573443e-05,
1667
+ "loss": 0.0562,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 8.880597014925373,
1672
+ "grad_norm": 0.20833946764469147,
1673
+ "learning_rate": 9.870399824239117e-05,
1674
+ "loss": 0.05,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 8.917910447761194,
1679
+ "grad_norm": 0.34507372975349426,
1680
+ "learning_rate": 9.868523060417646e-05,
1681
+ "loss": 0.0613,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 8.955223880597014,
1686
+ "grad_norm": 0.32865110039711,
1687
+ "learning_rate": 9.86663298624003e-05,
1688
+ "loss": 0.0621,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 8.992537313432836,
1693
+ "grad_norm": 0.21305270493030548,
1694
+ "learning_rate": 9.864729606873663e-05,
1695
+ "loss": 0.0572,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 9.029850746268657,
1700
+ "grad_norm": 0.28193730115890503,
1701
+ "learning_rate": 9.862812927522309e-05,
1702
+ "loss": 0.0555,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 9.067164179104477,
1707
+ "grad_norm": 0.3953789472579956,
1708
+ "learning_rate": 9.860882953426099e-05,
1709
+ "loss": 0.0536,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 9.104477611940299,
1714
+ "grad_norm": 0.23013322055339813,
1715
+ "learning_rate": 9.858939689861506e-05,
1716
+ "loss": 0.0572,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 9.14179104477612,
1721
+ "grad_norm": 0.2906680107116699,
1722
+ "learning_rate": 9.856983142141339e-05,
1723
+ "loss": 0.0592,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 9.17910447761194,
1728
+ "grad_norm": 0.23490828275680542,
1729
+ "learning_rate": 9.855013315614725e-05,
1730
+ "loss": 0.0583,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 9.216417910447761,
1735
+ "grad_norm": 0.22825880348682404,
1736
+ "learning_rate": 9.853030215667093e-05,
1737
+ "loss": 0.059,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 9.253731343283581,
1742
+ "grad_norm": 0.25871285796165466,
1743
+ "learning_rate": 9.851033847720166e-05,
1744
+ "loss": 0.0555,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 9.291044776119403,
1749
+ "grad_norm": 0.27220776677131653,
1750
+ "learning_rate": 9.849024217231935e-05,
1751
+ "loss": 0.0542,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 9.328358208955224,
1756
+ "grad_norm": 0.26534005999565125,
1757
+ "learning_rate": 9.847001329696653e-05,
1758
+ "loss": 0.0526,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 9.365671641791044,
1763
+ "grad_norm": 0.33486032485961914,
1764
+ "learning_rate": 9.844965190644817e-05,
1765
+ "loss": 0.0563,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 9.402985074626866,
1770
+ "grad_norm": 0.2949483394622803,
1771
+ "learning_rate": 9.842915805643155e-05,
1772
+ "loss": 0.0556,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 9.440298507462687,
1777
+ "grad_norm": 0.24123981595039368,
1778
+ "learning_rate": 9.840853180294608e-05,
1779
+ "loss": 0.05,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 9.477611940298507,
1784
+ "grad_norm": 0.22536049783229828,
1785
+ "learning_rate": 9.838777320238312e-05,
1786
+ "loss": 0.0522,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 9.514925373134329,
1791
+ "grad_norm": 0.23206663131713867,
1792
+ "learning_rate": 9.836688231149592e-05,
1793
+ "loss": 0.0591,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 9.552238805970148,
1798
+ "grad_norm": 0.28573134541511536,
1799
+ "learning_rate": 9.834585918739936e-05,
1800
+ "loss": 0.0568,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 9.58955223880597,
1805
+ "grad_norm": 0.2628820538520813,
1806
+ "learning_rate": 9.832470388756987e-05,
1807
+ "loss": 0.0571,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 9.626865671641792,
1812
+ "grad_norm": 0.2880440652370453,
1813
+ "learning_rate": 9.830341646984521e-05,
1814
+ "loss": 0.0559,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 9.664179104477611,
1819
+ "grad_norm": 0.1786259263753891,
1820
+ "learning_rate": 9.82819969924244e-05,
1821
+ "loss": 0.058,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 9.701492537313433,
1826
+ "grad_norm": 0.3501608073711395,
1827
+ "learning_rate": 9.826044551386744e-05,
1828
+ "loss": 0.0523,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 9.738805970149254,
1833
+ "grad_norm": 0.24757252633571625,
1834
+ "learning_rate": 9.823876209309527e-05,
1835
+ "loss": 0.0587,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 9.776119402985074,
1840
+ "grad_norm": 0.2556290626525879,
1841
+ "learning_rate": 9.821694678938953e-05,
1842
+ "loss": 0.0555,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 9.813432835820896,
1847
+ "grad_norm": 0.2561217248439789,
1848
+ "learning_rate": 9.819499966239243e-05,
1849
+ "loss": 0.052,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 9.850746268656717,
1854
+ "grad_norm": 0.2776634097099304,
1855
+ "learning_rate": 9.817292077210659e-05,
1856
+ "loss": 0.0498,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 9.888059701492537,
1861
+ "grad_norm": 0.20668549835681915,
1862
+ "learning_rate": 9.815071017889482e-05,
1863
+ "loss": 0.0517,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 9.925373134328359,
1868
+ "grad_norm": 0.3100263178348541,
1869
+ "learning_rate": 9.812836794348004e-05,
1870
+ "loss": 0.0633,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 9.962686567164178,
1875
+ "grad_norm": 0.2780782878398895,
1876
+ "learning_rate": 9.81058941269451e-05,
1877
+ "loss": 0.0581,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 10.0,
1882
+ "grad_norm": 0.28903728723526,
1883
+ "learning_rate": 9.808328879073251e-05,
1884
+ "loss": 0.0538,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 10.037313432835822,
1889
+ "grad_norm": 0.22727562487125397,
1890
+ "learning_rate": 9.806055199664446e-05,
1891
+ "loss": 0.0491,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 10.074626865671641,
1896
+ "grad_norm": 0.267918199300766,
1897
+ "learning_rate": 9.803768380684242e-05,
1898
+ "loss": 0.0562,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 10.111940298507463,
1903
+ "grad_norm": 0.2988606095314026,
1904
+ "learning_rate": 9.801468428384716e-05,
1905
+ "loss": 0.0566,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 10.149253731343283,
1910
+ "grad_norm": 0.2710281312465668,
1911
+ "learning_rate": 9.799155349053851e-05,
1912
+ "loss": 0.0541,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 10.186567164179104,
1917
+ "grad_norm": 0.15320520102977753,
1918
+ "learning_rate": 9.796829149015517e-05,
1919
+ "loss": 0.0548,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 10.223880597014926,
1924
+ "grad_norm": 0.2653089463710785,
1925
+ "learning_rate": 9.794489834629455e-05,
1926
+ "loss": 0.0599,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 10.261194029850746,
1931
+ "grad_norm": 0.19223959743976593,
1932
+ "learning_rate": 9.792137412291265e-05,
1933
+ "loss": 0.0494,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 10.298507462686567,
1938
+ "grad_norm": 0.20455987751483917,
1939
+ "learning_rate": 9.789771888432375e-05,
1940
+ "loss": 0.0538,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 10.335820895522389,
1945
+ "grad_norm": 0.24908749759197235,
1946
+ "learning_rate": 9.787393269520039e-05,
1947
+ "loss": 0.0481,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 10.373134328358208,
1952
+ "grad_norm": 0.3131813406944275,
1953
+ "learning_rate": 9.785001562057309e-05,
1954
+ "loss": 0.0526,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 10.41044776119403,
1959
+ "grad_norm": 0.24828971922397614,
1960
+ "learning_rate": 9.782596772583026e-05,
1961
+ "loss": 0.0489,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 10.447761194029852,
1966
+ "grad_norm": 0.21727119386196136,
1967
+ "learning_rate": 9.780178907671789e-05,
1968
+ "loss": 0.0532,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 10.485074626865671,
1973
+ "grad_norm": 0.20279547572135925,
1974
+ "learning_rate": 9.777747973933948e-05,
1975
+ "loss": 0.0565,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 10.522388059701493,
1980
+ "grad_norm": 0.17726702988147736,
1981
+ "learning_rate": 9.775303978015585e-05,
1982
+ "loss": 0.0437,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 10.559701492537313,
1987
+ "grad_norm": 0.18961119651794434,
1988
+ "learning_rate": 9.772846926598491e-05,
1989
+ "loss": 0.0584,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 10.597014925373134,
1994
+ "grad_norm": 0.2498980015516281,
1995
+ "learning_rate": 9.77037682640015e-05,
1996
+ "loss": 0.0496,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 10.634328358208956,
2001
+ "grad_norm": 0.16978798806667328,
2002
+ "learning_rate": 9.767893684173721e-05,
2003
+ "loss": 0.0469,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 10.671641791044776,
2008
+ "grad_norm": 0.16128584742546082,
2009
+ "learning_rate": 9.765397506708023e-05,
2010
+ "loss": 0.0533,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 10.708955223880597,
2015
+ "grad_norm": 0.20463155210018158,
2016
+ "learning_rate": 9.762888300827507e-05,
2017
+ "loss": 0.0464,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 10.746268656716419,
2022
+ "grad_norm": 0.30601629614830017,
2023
+ "learning_rate": 9.760366073392246e-05,
2024
+ "loss": 0.0489,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 10.783582089552239,
2029
+ "grad_norm": 0.2730671763420105,
2030
+ "learning_rate": 9.757830831297914e-05,
2031
+ "loss": 0.0495,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 10.82089552238806,
2036
+ "grad_norm": 0.251432865858078,
2037
+ "learning_rate": 9.755282581475769e-05,
2038
+ "loss": 0.0549,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 10.85820895522388,
2043
+ "grad_norm": 0.26670166850090027,
2044
+ "learning_rate": 9.752721330892624e-05,
2045
+ "loss": 0.061,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 10.895522388059701,
2050
+ "grad_norm": 0.2965967655181885,
2051
+ "learning_rate": 9.750147086550844e-05,
2052
+ "loss": 0.0473,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 10.932835820895523,
2057
+ "grad_norm": 0.683840274810791,
2058
+ "learning_rate": 9.747559855488313e-05,
2059
+ "loss": 0.0509,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 10.970149253731343,
2064
+ "grad_norm": 0.25740495324134827,
2065
+ "learning_rate": 9.744959644778422e-05,
2066
+ "loss": 0.0515,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 11.007462686567164,
2071
+ "grad_norm": 0.2880542278289795,
2072
+ "learning_rate": 9.742346461530048e-05,
2073
+ "loss": 0.0482,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 11.044776119402986,
2078
+ "grad_norm": 0.45032551884651184,
2079
+ "learning_rate": 9.739720312887535e-05,
2080
+ "loss": 0.0557,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 11.082089552238806,
2085
+ "grad_norm": 0.2829900085926056,
2086
+ "learning_rate": 9.73708120603067e-05,
2087
+ "loss": 0.052,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 11.119402985074627,
2092
+ "grad_norm": 0.309597373008728,
2093
+ "learning_rate": 9.734429148174675e-05,
2094
+ "loss": 0.0541,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 11.156716417910447,
2099
+ "grad_norm": 0.2433389127254486,
2100
+ "learning_rate": 9.731764146570173e-05,
2101
+ "loss": 0.0482,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 11.194029850746269,
2106
+ "grad_norm": 0.24458132684230804,
2107
+ "learning_rate": 9.729086208503174e-05,
2108
+ "loss": 0.0505,
2109
+ "step": 3000
2110
+ },
2111
+ {
2112
+ "epoch": 11.23134328358209,
2113
+ "grad_norm": 0.2305087298154831,
2114
+ "learning_rate": 9.726395341295062e-05,
2115
+ "loss": 0.0504,
2116
+ "step": 3010
2117
+ },
2118
+ {
2119
+ "epoch": 11.26865671641791,
2120
+ "grad_norm": 0.18110457062721252,
2121
+ "learning_rate": 9.723691552302562e-05,
2122
+ "loss": 0.0575,
2123
+ "step": 3020
2124
+ },
2125
+ {
2126
+ "epoch": 11.305970149253731,
2127
+ "grad_norm": 0.20407621562480927,
2128
+ "learning_rate": 9.720974848917735e-05,
2129
+ "loss": 0.0494,
2130
+ "step": 3030
2131
+ },
2132
+ {
2133
+ "epoch": 11.343283582089553,
2134
+ "grad_norm": 0.25924697518348694,
2135
+ "learning_rate": 9.718245238567939e-05,
2136
+ "loss": 0.0472,
2137
+ "step": 3040
2138
+ },
2139
+ {
2140
+ "epoch": 11.380597014925373,
2141
+ "grad_norm": 0.23041822016239166,
2142
+ "learning_rate": 9.715502728715826e-05,
2143
+ "loss": 0.0481,
2144
+ "step": 3050
2145
+ },
2146
+ {
2147
+ "epoch": 11.417910447761194,
2148
+ "grad_norm": 0.25381171703338623,
2149
+ "learning_rate": 9.712747326859315e-05,
2150
+ "loss": 0.0543,
2151
+ "step": 3060
2152
+ },
2153
+ {
2154
+ "epoch": 11.455223880597014,
2155
+ "grad_norm": 0.18027640879154205,
2156
+ "learning_rate": 9.709979040531569e-05,
2157
+ "loss": 0.055,
2158
+ "step": 3070
2159
+ },
2160
+ {
2161
+ "epoch": 11.492537313432836,
2162
+ "grad_norm": 0.2954868674278259,
2163
+ "learning_rate": 9.707197877300974e-05,
2164
+ "loss": 0.0473,
2165
+ "step": 3080
2166
+ },
2167
+ {
2168
+ "epoch": 11.529850746268657,
2169
+ "grad_norm": 0.25323861837387085,
2170
+ "learning_rate": 9.704403844771128e-05,
2171
+ "loss": 0.0509,
2172
+ "step": 3090
2173
+ },
2174
+ {
2175
+ "epoch": 11.567164179104477,
2176
+ "grad_norm": 0.36910176277160645,
2177
+ "learning_rate": 9.701596950580806e-05,
2178
+ "loss": 0.0504,
2179
+ "step": 3100
2180
+ },
2181
+ {
2182
+ "epoch": 11.604477611940299,
2183
+ "grad_norm": 0.34199246764183044,
2184
+ "learning_rate": 9.698777202403953e-05,
2185
+ "loss": 0.0526,
2186
+ "step": 3110
2187
+ },
2188
+ {
2189
+ "epoch": 11.64179104477612,
2190
+ "grad_norm": 0.2146557718515396,
2191
+ "learning_rate": 9.695944607949649e-05,
2192
+ "loss": 0.0579,
2193
+ "step": 3120
2194
+ },
2195
+ {
2196
+ "epoch": 11.67910447761194,
2197
+ "grad_norm": 0.20559175312519073,
2198
+ "learning_rate": 9.693099174962103e-05,
2199
+ "loss": 0.0514,
2200
+ "step": 3130
2201
+ },
2202
+ {
2203
+ "epoch": 11.716417910447761,
2204
+ "grad_norm": 0.2689419090747833,
2205
+ "learning_rate": 9.690240911220618e-05,
2206
+ "loss": 0.0534,
2207
+ "step": 3140
2208
+ },
2209
+ {
2210
+ "epoch": 11.753731343283581,
2211
+ "grad_norm": 0.34870603680610657,
2212
+ "learning_rate": 9.687369824539577e-05,
2213
+ "loss": 0.0485,
2214
+ "step": 3150
2215
+ },
2216
+ {
2217
+ "epoch": 11.791044776119403,
2218
+ "grad_norm": 0.15433363616466522,
2219
+ "learning_rate": 9.684485922768422e-05,
2220
+ "loss": 0.0418,
2221
+ "step": 3160
2222
+ },
2223
+ {
2224
+ "epoch": 11.828358208955224,
2225
+ "grad_norm": 0.26874423027038574,
2226
+ "learning_rate": 9.681589213791633e-05,
2227
+ "loss": 0.0537,
2228
+ "step": 3170
2229
+ },
2230
+ {
2231
+ "epoch": 11.865671641791044,
2232
+ "grad_norm": 0.3361654281616211,
2233
+ "learning_rate": 9.6786797055287e-05,
2234
+ "loss": 0.0474,
2235
+ "step": 3180
2236
+ },
2237
+ {
2238
+ "epoch": 11.902985074626866,
2239
+ "grad_norm": 0.17938771843910217,
2240
+ "learning_rate": 9.675757405934103e-05,
2241
+ "loss": 0.0443,
2242
+ "step": 3190
2243
+ },
2244
+ {
2245
+ "epoch": 11.940298507462687,
2246
+ "grad_norm": 0.31368622183799744,
2247
+ "learning_rate": 9.672822322997305e-05,
2248
+ "loss": 0.0594,
2249
+ "step": 3200
2250
+ },
2251
+ {
2252
+ "epoch": 11.977611940298507,
2253
+ "grad_norm": 0.16268151998519897,
2254
+ "learning_rate": 9.669874464742705e-05,
2255
+ "loss": 0.0487,
2256
+ "step": 3210
2257
+ },
2258
+ {
2259
+ "epoch": 12.014925373134329,
2260
+ "grad_norm": 0.23879969120025635,
2261
+ "learning_rate": 9.66691383922964e-05,
2262
+ "loss": 0.0484,
2263
+ "step": 3220
2264
+ },
2265
+ {
2266
+ "epoch": 12.052238805970148,
2267
+ "grad_norm": 0.2321789413690567,
2268
+ "learning_rate": 9.663940454552342e-05,
2269
+ "loss": 0.051,
2270
+ "step": 3230
2271
+ },
2272
+ {
2273
+ "epoch": 12.08955223880597,
2274
+ "grad_norm": 0.22873088717460632,
2275
+ "learning_rate": 9.660954318839933e-05,
2276
+ "loss": 0.0406,
2277
+ "step": 3240
2278
+ },
2279
+ {
2280
+ "epoch": 12.126865671641792,
2281
+ "grad_norm": 0.3767557740211487,
2282
+ "learning_rate": 9.657955440256395e-05,
2283
+ "loss": 0.0432,
2284
+ "step": 3250
2285
+ },
2286
+ {
2287
+ "epoch": 12.164179104477611,
2288
+ "grad_norm": 0.21569453179836273,
2289
+ "learning_rate": 9.654943827000548e-05,
2290
+ "loss": 0.0528,
2291
+ "step": 3260
2292
+ },
2293
+ {
2294
+ "epoch": 12.201492537313433,
2295
+ "grad_norm": 0.23698291182518005,
2296
+ "learning_rate": 9.651919487306025e-05,
2297
+ "loss": 0.0457,
2298
+ "step": 3270
2299
+ },
2300
+ {
2301
+ "epoch": 12.238805970149254,
2302
+ "grad_norm": 0.21086478233337402,
2303
+ "learning_rate": 9.648882429441257e-05,
2304
+ "loss": 0.0508,
2305
+ "step": 3280
2306
+ },
2307
+ {
2308
+ "epoch": 12.276119402985074,
2309
+ "grad_norm": 0.19763463735580444,
2310
+ "learning_rate": 9.645832661709444e-05,
2311
+ "loss": 0.0497,
2312
+ "step": 3290
2313
+ },
2314
+ {
2315
+ "epoch": 12.313432835820896,
2316
+ "grad_norm": 0.18413852155208588,
2317
+ "learning_rate": 9.642770192448536e-05,
2318
+ "loss": 0.0441,
2319
+ "step": 3300
2320
+ },
2321
+ {
2322
+ "epoch": 12.350746268656717,
2323
+ "grad_norm": 0.13946911692619324,
2324
+ "learning_rate": 9.639695030031204e-05,
2325
+ "loss": 0.0453,
2326
+ "step": 3310
2327
+ },
2328
+ {
2329
+ "epoch": 12.388059701492537,
2330
+ "grad_norm": 0.21613670885562897,
2331
+ "learning_rate": 9.636607182864827e-05,
2332
+ "loss": 0.0511,
2333
+ "step": 3320
2334
+ },
2335
+ {
2336
+ "epoch": 12.425373134328359,
2337
+ "grad_norm": 0.24953646957874298,
2338
+ "learning_rate": 9.63350665939146e-05,
2339
+ "loss": 0.0451,
2340
+ "step": 3330
2341
+ },
2342
+ {
2343
+ "epoch": 12.462686567164178,
2344
+ "grad_norm": 0.2993795871734619,
2345
+ "learning_rate": 9.630393468087818e-05,
2346
+ "loss": 0.0469,
2347
+ "step": 3340
2348
+ },
2349
+ {
2350
+ "epoch": 12.5,
2351
+ "grad_norm": 0.2261819839477539,
2352
+ "learning_rate": 9.627267617465243e-05,
2353
+ "loss": 0.0484,
2354
+ "step": 3350
2355
+ },
2356
+ {
2357
+ "epoch": 12.537313432835822,
2358
+ "grad_norm": 0.23026186227798462,
2359
+ "learning_rate": 9.624129116069694e-05,
2360
+ "loss": 0.0452,
2361
+ "step": 3360
2362
+ },
2363
+ {
2364
+ "epoch": 12.574626865671641,
2365
+ "grad_norm": 0.27859947085380554,
2366
+ "learning_rate": 9.620977972481716e-05,
2367
+ "loss": 0.0593,
2368
+ "step": 3370
2369
+ },
2370
+ {
2371
+ "epoch": 12.611940298507463,
2372
+ "grad_norm": 0.23060785233974457,
2373
+ "learning_rate": 9.617814195316411e-05,
2374
+ "loss": 0.05,
2375
+ "step": 3380
2376
+ },
2377
+ {
2378
+ "epoch": 12.649253731343283,
2379
+ "grad_norm": 0.20185025036334991,
2380
+ "learning_rate": 9.614637793223425e-05,
2381
+ "loss": 0.0573,
2382
+ "step": 3390
2383
+ },
2384
+ {
2385
+ "epoch": 12.686567164179104,
2386
+ "grad_norm": 0.3584498167037964,
2387
+ "learning_rate": 9.611448774886924e-05,
2388
+ "loss": 0.052,
2389
+ "step": 3400
2390
+ },
2391
+ {
2392
+ "epoch": 12.723880597014926,
2393
+ "grad_norm": 0.19336827099323273,
2394
+ "learning_rate": 9.60824714902556e-05,
2395
+ "loss": 0.0535,
2396
+ "step": 3410
2397
+ },
2398
+ {
2399
+ "epoch": 12.761194029850746,
2400
+ "grad_norm": 0.22223635017871857,
2401
+ "learning_rate": 9.605032924392457e-05,
2402
+ "loss": 0.05,
2403
+ "step": 3420
2404
+ },
2405
+ {
2406
+ "epoch": 12.798507462686567,
2407
+ "grad_norm": 0.17108851671218872,
2408
+ "learning_rate": 9.601806109775179e-05,
2409
+ "loss": 0.0475,
2410
+ "step": 3430
2411
+ },
2412
+ {
2413
+ "epoch": 12.835820895522389,
2414
+ "grad_norm": 0.3861902952194214,
2415
+ "learning_rate": 9.598566713995718e-05,
2416
+ "loss": 0.0439,
2417
+ "step": 3440
2418
+ },
2419
+ {
2420
+ "epoch": 12.873134328358208,
2421
+ "grad_norm": 0.18927253782749176,
2422
+ "learning_rate": 9.595314745910456e-05,
2423
+ "loss": 0.052,
2424
+ "step": 3450
2425
+ },
2426
+ {
2427
+ "epoch": 12.91044776119403,
2428
+ "grad_norm": 0.21963383257389069,
2429
+ "learning_rate": 9.59205021441015e-05,
2430
+ "loss": 0.0504,
2431
+ "step": 3460
2432
+ },
2433
+ {
2434
+ "epoch": 12.947761194029852,
2435
+ "grad_norm": 0.18016670644283295,
2436
+ "learning_rate": 9.588773128419906e-05,
2437
+ "loss": 0.0467,
2438
+ "step": 3470
2439
+ },
2440
+ {
2441
+ "epoch": 12.985074626865671,
2442
+ "grad_norm": 0.1776365041732788,
2443
+ "learning_rate": 9.58548349689915e-05,
2444
+ "loss": 0.0414,
2445
+ "step": 3480
2446
+ },
2447
+ {
2448
+ "epoch": 13.022388059701493,
2449
+ "grad_norm": 0.2616482973098755,
2450
+ "learning_rate": 9.582181328841611e-05,
2451
+ "loss": 0.0442,
2452
+ "step": 3490
2453
+ },
2454
+ {
2455
+ "epoch": 13.059701492537313,
2456
+ "grad_norm": 0.20341171324253082,
2457
+ "learning_rate": 9.578866633275288e-05,
2458
+ "loss": 0.0533,
2459
+ "step": 3500
2460
+ },
2461
+ {
2462
+ "epoch": 13.097014925373134,
2463
+ "grad_norm": 0.2223699688911438,
2464
+ "learning_rate": 9.575539419262434e-05,
2465
+ "loss": 0.0458,
2466
+ "step": 3510
2467
+ },
2468
+ {
2469
+ "epoch": 13.134328358208956,
2470
+ "grad_norm": 0.22557464241981506,
2471
+ "learning_rate": 9.572199695899522e-05,
2472
+ "loss": 0.0445,
2473
+ "step": 3520
2474
+ },
2475
+ {
2476
+ "epoch": 13.171641791044776,
2477
+ "grad_norm": 0.25104308128356934,
2478
+ "learning_rate": 9.568847472317232e-05,
2479
+ "loss": 0.0435,
2480
+ "step": 3530
2481
+ },
2482
+ {
2483
+ "epoch": 13.208955223880597,
2484
+ "grad_norm": 0.18720711767673492,
2485
+ "learning_rate": 9.565482757680415e-05,
2486
+ "loss": 0.0453,
2487
+ "step": 3540
2488
+ },
2489
+ {
2490
+ "epoch": 13.246268656716419,
2491
+ "grad_norm": 0.16838951408863068,
2492
+ "learning_rate": 9.562105561188069e-05,
2493
+ "loss": 0.0505,
2494
+ "step": 3550
2495
+ },
2496
+ {
2497
+ "epoch": 13.283582089552239,
2498
+ "grad_norm": 0.31681734323501587,
2499
+ "learning_rate": 9.558715892073323e-05,
2500
+ "loss": 0.0494,
2501
+ "step": 3560
2502
+ },
2503
+ {
2504
+ "epoch": 13.32089552238806,
2505
+ "grad_norm": 0.2390700727701187,
2506
+ "learning_rate": 9.555313759603402e-05,
2507
+ "loss": 0.0538,
2508
+ "step": 3570
2509
+ },
2510
+ {
2511
+ "epoch": 13.35820895522388,
2512
+ "grad_norm": 0.20680709183216095,
2513
+ "learning_rate": 9.551899173079607e-05,
2514
+ "loss": 0.0519,
2515
+ "step": 3580
2516
+ },
2517
+ {
2518
+ "epoch": 13.395522388059701,
2519
+ "grad_norm": 0.2758580148220062,
2520
+ "learning_rate": 9.548472141837286e-05,
2521
+ "loss": 0.0512,
2522
+ "step": 3590
2523
+ },
2524
+ {
2525
+ "epoch": 13.432835820895523,
2526
+ "grad_norm": 0.3653097450733185,
2527
+ "learning_rate": 9.545032675245813e-05,
2528
+ "loss": 0.0496,
2529
+ "step": 3600
2530
+ },
2531
+ {
2532
+ "epoch": 13.470149253731343,
2533
+ "grad_norm": 0.23886866867542267,
2534
+ "learning_rate": 9.541580782708557e-05,
2535
+ "loss": 0.0455,
2536
+ "step": 3610
2537
+ },
2538
+ {
2539
+ "epoch": 13.507462686567164,
2540
+ "grad_norm": 0.3280908465385437,
2541
+ "learning_rate": 9.538116473662861e-05,
2542
+ "loss": 0.0489,
2543
+ "step": 3620
2544
+ },
2545
+ {
2546
+ "epoch": 13.544776119402986,
2547
+ "grad_norm": 0.20268180966377258,
2548
+ "learning_rate": 9.534639757580013e-05,
2549
+ "loss": 0.0484,
2550
+ "step": 3630
2551
+ },
2552
+ {
2553
+ "epoch": 13.582089552238806,
2554
+ "grad_norm": 0.2582015097141266,
2555
+ "learning_rate": 9.531150643965223e-05,
2556
+ "loss": 0.0487,
2557
+ "step": 3640
2558
+ },
2559
+ {
2560
+ "epoch": 13.619402985074627,
2561
+ "grad_norm": 0.18157973885536194,
2562
+ "learning_rate": 9.527649142357596e-05,
2563
+ "loss": 0.0496,
2564
+ "step": 3650
2565
+ },
2566
+ {
2567
+ "epoch": 13.656716417910447,
2568
+ "grad_norm": 0.22841542959213257,
2569
+ "learning_rate": 9.524135262330098e-05,
2570
+ "loss": 0.0467,
2571
+ "step": 3660
2572
+ },
2573
+ {
2574
+ "epoch": 13.694029850746269,
2575
+ "grad_norm": 0.2519935369491577,
2576
+ "learning_rate": 9.520609013489547e-05,
2577
+ "loss": 0.0487,
2578
+ "step": 3670
2579
+ },
2580
+ {
2581
+ "epoch": 13.73134328358209,
2582
+ "grad_norm": 0.24680495262145996,
2583
+ "learning_rate": 9.517070405476575e-05,
2584
+ "loss": 0.0457,
2585
+ "step": 3680
2586
+ },
2587
+ {
2588
+ "epoch": 13.76865671641791,
2589
+ "grad_norm": 0.26362067461013794,
2590
+ "learning_rate": 9.513519447965595e-05,
2591
+ "loss": 0.0495,
2592
+ "step": 3690
2593
+ },
2594
+ {
2595
+ "epoch": 13.805970149253731,
2596
+ "grad_norm": 0.3240712583065033,
2597
+ "learning_rate": 9.509956150664796e-05,
2598
+ "loss": 0.0496,
2599
+ "step": 3700
2600
+ },
2601
+ {
2602
+ "epoch": 13.843283582089553,
2603
+ "grad_norm": 0.21009013056755066,
2604
+ "learning_rate": 9.50638052331609e-05,
2605
+ "loss": 0.0457,
2606
+ "step": 3710
2607
+ },
2608
+ {
2609
+ "epoch": 13.880597014925373,
2610
+ "grad_norm": 0.1669154316186905,
2611
+ "learning_rate": 9.502792575695112e-05,
2612
+ "loss": 0.0496,
2613
+ "step": 3720
2614
+ },
2615
+ {
2616
+ "epoch": 13.917910447761194,
2617
+ "grad_norm": 0.22347605228424072,
2618
+ "learning_rate": 9.499192317611167e-05,
2619
+ "loss": 0.0426,
2620
+ "step": 3730
2621
+ },
2622
+ {
2623
+ "epoch": 13.955223880597014,
2624
+ "grad_norm": 0.15208907425403595,
2625
+ "learning_rate": 9.49557975890723e-05,
2626
+ "loss": 0.0447,
2627
+ "step": 3740
2628
+ },
2629
+ {
2630
+ "epoch": 13.992537313432836,
2631
+ "grad_norm": 0.3206101059913635,
2632
+ "learning_rate": 9.491954909459895e-05,
2633
+ "loss": 0.0471,
2634
+ "step": 3750
2635
+ },
2636
+ {
2637
+ "epoch": 14.029850746268657,
2638
+ "grad_norm": 0.15873713791370392,
2639
+ "learning_rate": 9.488317779179361e-05,
2640
+ "loss": 0.0401,
2641
+ "step": 3760
2642
+ },
2643
+ {
2644
+ "epoch": 14.067164179104477,
2645
+ "grad_norm": 0.19690357148647308,
2646
+ "learning_rate": 9.484668378009408e-05,
2647
+ "loss": 0.0491,
2648
+ "step": 3770
2649
+ },
2650
+ {
2651
+ "epoch": 14.104477611940299,
2652
+ "grad_norm": 0.3211113214492798,
2653
+ "learning_rate": 9.481006715927351e-05,
2654
+ "loss": 0.049,
2655
+ "step": 3780
2656
+ },
2657
+ {
2658
+ "epoch": 14.14179104477612,
2659
+ "grad_norm": 0.27657604217529297,
2660
+ "learning_rate": 9.477332802944044e-05,
2661
+ "loss": 0.0396,
2662
+ "step": 3790
2663
+ },
2664
+ {
2665
+ "epoch": 14.17910447761194,
2666
+ "grad_norm": 0.20194031298160553,
2667
+ "learning_rate": 9.473646649103818e-05,
2668
+ "loss": 0.0442,
2669
+ "step": 3800
2670
+ },
2671
+ {
2672
+ "epoch": 14.216417910447761,
2673
+ "grad_norm": 0.20344595611095428,
2674
+ "learning_rate": 9.46994826448448e-05,
2675
+ "loss": 0.0427,
2676
+ "step": 3810
2677
+ },
2678
+ {
2679
+ "epoch": 14.253731343283581,
2680
+ "grad_norm": 0.2067718505859375,
2681
+ "learning_rate": 9.46623765919727e-05,
2682
+ "loss": 0.0501,
2683
+ "step": 3820
2684
+ },
2685
+ {
2686
+ "epoch": 14.291044776119403,
2687
+ "grad_norm": 0.29719170928001404,
2688
+ "learning_rate": 9.462514843386845e-05,
2689
+ "loss": 0.0519,
2690
+ "step": 3830
2691
+ },
2692
+ {
2693
+ "epoch": 14.328358208955224,
2694
+ "grad_norm": 0.2347182184457779,
2695
+ "learning_rate": 9.458779827231237e-05,
2696
+ "loss": 0.0413,
2697
+ "step": 3840
2698
+ },
2699
+ {
2700
+ "epoch": 14.365671641791044,
2701
+ "grad_norm": 0.1558852344751358,
2702
+ "learning_rate": 9.45503262094184e-05,
2703
+ "loss": 0.0442,
2704
+ "step": 3850
2705
+ },
2706
+ {
2707
+ "epoch": 14.402985074626866,
2708
+ "grad_norm": 0.23085005581378937,
2709
+ "learning_rate": 9.451273234763371e-05,
2710
+ "loss": 0.047,
2711
+ "step": 3860
2712
+ },
2713
+ {
2714
+ "epoch": 14.440298507462687,
2715
+ "grad_norm": 0.1515151560306549,
2716
+ "learning_rate": 9.447501678973852e-05,
2717
+ "loss": 0.0481,
2718
+ "step": 3870
2719
+ },
2720
+ {
2721
+ "epoch": 14.477611940298507,
2722
+ "grad_norm": 0.1916729211807251,
2723
+ "learning_rate": 9.443717963884569e-05,
2724
+ "loss": 0.0474,
2725
+ "step": 3880
2726
+ },
2727
+ {
2728
+ "epoch": 14.514925373134329,
2729
+ "grad_norm": 0.2536492943763733,
2730
+ "learning_rate": 9.439922099840054e-05,
2731
+ "loss": 0.0382,
2732
+ "step": 3890
2733
+ },
2734
+ {
2735
+ "epoch": 14.552238805970148,
2736
+ "grad_norm": 0.1672086864709854,
2737
+ "learning_rate": 9.43611409721806e-05,
2738
+ "loss": 0.0497,
2739
+ "step": 3900
2740
+ },
2741
+ {
2742
+ "epoch": 14.58955223880597,
2743
+ "grad_norm": 0.3644237518310547,
2744
+ "learning_rate": 9.432293966429514e-05,
2745
+ "loss": 0.0444,
2746
+ "step": 3910
2747
+ },
2748
+ {
2749
+ "epoch": 14.626865671641792,
2750
+ "grad_norm": 0.20307251811027527,
2751
+ "learning_rate": 9.428461717918511e-05,
2752
+ "loss": 0.0452,
2753
+ "step": 3920
2754
+ },
2755
+ {
2756
+ "epoch": 14.664179104477611,
2757
+ "grad_norm": 0.20441733300685883,
2758
+ "learning_rate": 9.424617362162271e-05,
2759
+ "loss": 0.0454,
2760
+ "step": 3930
2761
+ },
2762
+ {
2763
+ "epoch": 14.701492537313433,
2764
+ "grad_norm": 0.26315611600875854,
2765
+ "learning_rate": 9.420760909671118e-05,
2766
+ "loss": 0.0486,
2767
+ "step": 3940
2768
+ },
2769
+ {
2770
+ "epoch": 14.738805970149254,
2771
+ "grad_norm": 0.1983092874288559,
2772
+ "learning_rate": 9.416892370988444e-05,
2773
+ "loss": 0.0483,
2774
+ "step": 3950
2775
+ },
2776
+ {
2777
+ "epoch": 14.776119402985074,
2778
+ "grad_norm": 0.18301443755626678,
2779
+ "learning_rate": 9.413011756690685e-05,
2780
+ "loss": 0.0456,
2781
+ "step": 3960
2782
+ },
2783
+ {
2784
+ "epoch": 14.813432835820896,
2785
+ "grad_norm": 0.2433597594499588,
2786
+ "learning_rate": 9.409119077387294e-05,
2787
+ "loss": 0.0463,
2788
+ "step": 3970
2789
+ },
2790
+ {
2791
+ "epoch": 14.850746268656717,
2792
+ "grad_norm": 0.27949392795562744,
2793
+ "learning_rate": 9.405214343720707e-05,
2794
+ "loss": 0.0412,
2795
+ "step": 3980
2796
+ },
2797
+ {
2798
+ "epoch": 14.888059701492537,
2799
+ "grad_norm": 0.22806599736213684,
2800
+ "learning_rate": 9.401297566366318e-05,
2801
+ "loss": 0.0448,
2802
+ "step": 3990
2803
+ },
2804
+ {
2805
+ "epoch": 14.925373134328359,
2806
+ "grad_norm": 0.25421562790870667,
2807
+ "learning_rate": 9.397368756032445e-05,
2808
+ "loss": 0.0426,
2809
+ "step": 4000
2810
+ }
2811
+ ],
2812
+ "logging_steps": 10,
2813
+ "max_steps": 20000,
2814
+ "num_input_tokens_seen": 0,
2815
+ "num_train_epochs": 75,
2816
+ "save_steps": 500,
2817
+ "stateful_callbacks": {
2818
+ "TrainerControl": {
2819
+ "args": {
2820
+ "should_epoch_stop": false,
2821
+ "should_evaluate": false,
2822
+ "should_log": false,
2823
+ "should_save": true,
2824
+ "should_training_stop": false
2825
+ },
2826
+ "attributes": {}
2827
+ }
2828
+ },
2829
+ "total_flos": 3.0070179379263744e+17,
2830
+ "train_batch_size": 32,
2831
+ "trial_name": null,
2832
+ "trial_params": null
2833
+ }
checkpoint-4500/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-4500/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-4500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67195aed8359e845aa6e736c609a2360ef327f1e8d7a1e7603ea5bd0aed0f09c
3
+ size 123328576
checkpoint-4500/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-4500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7d3a2a860bfa94c86db53c686d2fcbaaf7931b0c992b1d60b8b794ee205fe85
3
+ size 246824634
checkpoint-4500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a9c3f7e9d7afb450f369a250eea2d964c6dd9b992e90e251ae8ac092e097545
3
+ size 14244
checkpoint-4500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26ded847c9c7706bb864044bf340d57043e09157b2acfb3ad0c66b2c6ae892c2
3
+ size 1064
checkpoint-4500/trainer_state.json ADDED
@@ -0,0 +1,3183 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 16.791044776119403,
5
+ "eval_steps": 500,
6
+ "global_step": 4500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03731343283582089,
13
+ "grad_norm": 0.8186072111129761,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 1.3847,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.07462686567164178,
20
+ "grad_norm": 0.5007426142692566,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 1.4283,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.11194029850746269,
27
+ "grad_norm": 0.49460887908935547,
28
+ "learning_rate": 3e-06,
29
+ "loss": 1.4868,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.14925373134328357,
34
+ "grad_norm": 0.5032920837402344,
35
+ "learning_rate": 4.000000000000001e-06,
36
+ "loss": 1.4491,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1865671641791045,
41
+ "grad_norm": 0.5688469409942627,
42
+ "learning_rate": 5e-06,
43
+ "loss": 1.3703,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.22388059701492538,
48
+ "grad_norm": 0.5052517652511597,
49
+ "learning_rate": 6e-06,
50
+ "loss": 1.419,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.26119402985074625,
55
+ "grad_norm": 0.6315643787384033,
56
+ "learning_rate": 7.000000000000001e-06,
57
+ "loss": 1.3058,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.29850746268656714,
62
+ "grad_norm": 0.6060447692871094,
63
+ "learning_rate": 8.000000000000001e-06,
64
+ "loss": 1.2908,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3358208955223881,
69
+ "grad_norm": 0.5513179302215576,
70
+ "learning_rate": 9e-06,
71
+ "loss": 1.2311,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.373134328358209,
76
+ "grad_norm": 0.8467404246330261,
77
+ "learning_rate": 1e-05,
78
+ "loss": 1.2043,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.41044776119402987,
83
+ "grad_norm": 0.8141824007034302,
84
+ "learning_rate": 1.1000000000000001e-05,
85
+ "loss": 1.0707,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.44776119402985076,
90
+ "grad_norm": 0.7932347059249878,
91
+ "learning_rate": 1.2e-05,
92
+ "loss": 0.9377,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.48507462686567165,
97
+ "grad_norm": 0.684220552444458,
98
+ "learning_rate": 1.3000000000000001e-05,
99
+ "loss": 0.714,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5223880597014925,
104
+ "grad_norm": 0.5886895060539246,
105
+ "learning_rate": 1.4000000000000001e-05,
106
+ "loss": 0.6479,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5597014925373134,
111
+ "grad_norm": 0.4764939248561859,
112
+ "learning_rate": 1.5e-05,
113
+ "loss": 0.5463,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.5970149253731343,
118
+ "grad_norm": 0.4621008038520813,
119
+ "learning_rate": 1.6000000000000003e-05,
120
+ "loss": 0.4641,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6343283582089553,
125
+ "grad_norm": 0.46492910385131836,
126
+ "learning_rate": 1.7000000000000003e-05,
127
+ "loss": 0.4159,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6716417910447762,
132
+ "grad_norm": 0.5017415881156921,
133
+ "learning_rate": 1.8e-05,
134
+ "loss": 0.4094,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7089552238805971,
139
+ "grad_norm": 0.34392210841178894,
140
+ "learning_rate": 1.9e-05,
141
+ "loss": 0.3478,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.746268656716418,
146
+ "grad_norm": 0.3240516483783722,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3821,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7835820895522388,
153
+ "grad_norm": 0.26301339268684387,
154
+ "learning_rate": 2.1e-05,
155
+ "loss": 0.3606,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8208955223880597,
160
+ "grad_norm": 0.34712520241737366,
161
+ "learning_rate": 2.2000000000000003e-05,
162
+ "loss": 0.3421,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8582089552238806,
167
+ "grad_norm": 0.3248469829559326,
168
+ "learning_rate": 2.3000000000000003e-05,
169
+ "loss": 0.3389,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.8955223880597015,
174
+ "grad_norm": 0.298149436712265,
175
+ "learning_rate": 2.4e-05,
176
+ "loss": 0.3145,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.9328358208955224,
181
+ "grad_norm": 0.2757190763950348,
182
+ "learning_rate": 2.5e-05,
183
+ "loss": 0.3065,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9701492537313433,
188
+ "grad_norm": 0.30510950088500977,
189
+ "learning_rate": 2.6000000000000002e-05,
190
+ "loss": 0.2971,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.007462686567164,
195
+ "grad_norm": 0.37349891662597656,
196
+ "learning_rate": 2.7000000000000002e-05,
197
+ "loss": 0.3273,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.044776119402985,
202
+ "grad_norm": 0.3667634129524231,
203
+ "learning_rate": 2.8000000000000003e-05,
204
+ "loss": 0.308,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.0820895522388059,
209
+ "grad_norm": 0.3463355004787445,
210
+ "learning_rate": 2.9e-05,
211
+ "loss": 0.3109,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.1194029850746268,
216
+ "grad_norm": 0.3888525366783142,
217
+ "learning_rate": 3e-05,
218
+ "loss": 0.2644,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.1567164179104479,
223
+ "grad_norm": 0.3749147951602936,
224
+ "learning_rate": 3.1e-05,
225
+ "loss": 0.2858,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.1940298507462686,
230
+ "grad_norm": 0.3270276188850403,
231
+ "learning_rate": 3.2000000000000005e-05,
232
+ "loss": 0.2573,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.2313432835820897,
237
+ "grad_norm": 0.3658592998981476,
238
+ "learning_rate": 3.3e-05,
239
+ "loss": 0.2613,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.2686567164179103,
244
+ "grad_norm": 0.3526328206062317,
245
+ "learning_rate": 3.4000000000000007e-05,
246
+ "loss": 0.2328,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.3059701492537314,
251
+ "grad_norm": 0.4528139531612396,
252
+ "learning_rate": 3.5e-05,
253
+ "loss": 0.2429,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.3432835820895521,
258
+ "grad_norm": 0.5426791310310364,
259
+ "learning_rate": 3.6e-05,
260
+ "loss": 0.2209,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.3805970149253732,
265
+ "grad_norm": 0.41844552755355835,
266
+ "learning_rate": 3.7e-05,
267
+ "loss": 0.2319,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.417910447761194,
272
+ "grad_norm": 0.4749431908130646,
273
+ "learning_rate": 3.8e-05,
274
+ "loss": 0.2233,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.455223880597015,
279
+ "grad_norm": 0.7010189890861511,
280
+ "learning_rate": 3.9000000000000006e-05,
281
+ "loss": 0.2181,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.4925373134328357,
286
+ "grad_norm": 0.5747635960578918,
287
+ "learning_rate": 4e-05,
288
+ "loss": 0.213,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.5298507462686568,
293
+ "grad_norm": 0.3661474287509918,
294
+ "learning_rate": 4.1e-05,
295
+ "loss": 0.2171,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.5671641791044775,
300
+ "grad_norm": 0.467835396528244,
301
+ "learning_rate": 4.2e-05,
302
+ "loss": 0.1985,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.6044776119402986,
307
+ "grad_norm": 0.5470123291015625,
308
+ "learning_rate": 4.3e-05,
309
+ "loss": 0.2176,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.6417910447761193,
314
+ "grad_norm": 0.5761199593544006,
315
+ "learning_rate": 4.4000000000000006e-05,
316
+ "loss": 0.2007,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.6791044776119404,
321
+ "grad_norm": 0.48257485032081604,
322
+ "learning_rate": 4.5e-05,
323
+ "loss": 0.2043,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.716417910447761,
328
+ "grad_norm": 0.48353052139282227,
329
+ "learning_rate": 4.600000000000001e-05,
330
+ "loss": 0.1872,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.7537313432835822,
335
+ "grad_norm": 0.4388391375541687,
336
+ "learning_rate": 4.7e-05,
337
+ "loss": 0.206,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.7910447761194028,
342
+ "grad_norm": 0.47332626581192017,
343
+ "learning_rate": 4.8e-05,
344
+ "loss": 0.1876,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.828358208955224,
349
+ "grad_norm": 0.8053535223007202,
350
+ "learning_rate": 4.9e-05,
351
+ "loss": 0.1839,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 1.8656716417910446,
356
+ "grad_norm": 0.413979709148407,
357
+ "learning_rate": 5e-05,
358
+ "loss": 0.1732,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 1.9029850746268657,
363
+ "grad_norm": 0.36910712718963623,
364
+ "learning_rate": 5.1000000000000006e-05,
365
+ "loss": 0.1827,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 1.9402985074626866,
370
+ "grad_norm": 0.8458298444747925,
371
+ "learning_rate": 5.2000000000000004e-05,
372
+ "loss": 0.1727,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 1.9776119402985075,
377
+ "grad_norm": 0.5452115535736084,
378
+ "learning_rate": 5.300000000000001e-05,
379
+ "loss": 0.1818,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.014925373134328,
384
+ "grad_norm": 0.4518108069896698,
385
+ "learning_rate": 5.4000000000000005e-05,
386
+ "loss": 0.177,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.0522388059701493,
391
+ "grad_norm": 0.66865074634552,
392
+ "learning_rate": 5.500000000000001e-05,
393
+ "loss": 0.1726,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.08955223880597,
398
+ "grad_norm": 0.6536034345626831,
399
+ "learning_rate": 5.6000000000000006e-05,
400
+ "loss": 0.1541,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.126865671641791,
405
+ "grad_norm": 0.5571377277374268,
406
+ "learning_rate": 5.6999999999999996e-05,
407
+ "loss": 0.1671,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.1641791044776117,
412
+ "grad_norm": 0.5385546684265137,
413
+ "learning_rate": 5.8e-05,
414
+ "loss": 0.1582,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.201492537313433,
419
+ "grad_norm": 0.577961266040802,
420
+ "learning_rate": 5.9e-05,
421
+ "loss": 0.1528,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.2388059701492535,
426
+ "grad_norm": 0.5082416534423828,
427
+ "learning_rate": 6e-05,
428
+ "loss": 0.1638,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.2761194029850746,
433
+ "grad_norm": 0.5490861535072327,
434
+ "learning_rate": 6.1e-05,
435
+ "loss": 0.166,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.3134328358208958,
440
+ "grad_norm": 0.492366760969162,
441
+ "learning_rate": 6.2e-05,
442
+ "loss": 0.1481,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.3507462686567164,
447
+ "grad_norm": 0.3702855110168457,
448
+ "learning_rate": 6.3e-05,
449
+ "loss": 0.1514,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.388059701492537,
454
+ "grad_norm": 0.664667010307312,
455
+ "learning_rate": 6.400000000000001e-05,
456
+ "loss": 0.1441,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.425373134328358,
461
+ "grad_norm": 0.33382174372673035,
462
+ "learning_rate": 6.500000000000001e-05,
463
+ "loss": 0.1573,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.4626865671641793,
468
+ "grad_norm": 0.4848814010620117,
469
+ "learning_rate": 6.6e-05,
470
+ "loss": 0.1457,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.5,
475
+ "grad_norm": 0.3649997413158417,
476
+ "learning_rate": 6.7e-05,
477
+ "loss": 0.1467,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.5373134328358207,
482
+ "grad_norm": 0.6385223865509033,
483
+ "learning_rate": 6.800000000000001e-05,
484
+ "loss": 0.145,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.574626865671642,
489
+ "grad_norm": 0.4580625891685486,
490
+ "learning_rate": 6.9e-05,
491
+ "loss": 0.1352,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.611940298507463,
496
+ "grad_norm": 0.5141746401786804,
497
+ "learning_rate": 7e-05,
498
+ "loss": 0.1444,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.6492537313432836,
503
+ "grad_norm": 0.40220722556114197,
504
+ "learning_rate": 7.1e-05,
505
+ "loss": 0.1493,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.6865671641791042,
510
+ "grad_norm": 0.5510571002960205,
511
+ "learning_rate": 7.2e-05,
512
+ "loss": 0.1387,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.7238805970149254,
517
+ "grad_norm": 0.43814659118652344,
518
+ "learning_rate": 7.3e-05,
519
+ "loss": 0.1374,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 2.7611940298507465,
524
+ "grad_norm": 0.4118008613586426,
525
+ "learning_rate": 7.4e-05,
526
+ "loss": 0.1297,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 2.798507462686567,
531
+ "grad_norm": 0.5626503229141235,
532
+ "learning_rate": 7.500000000000001e-05,
533
+ "loss": 0.1299,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 2.835820895522388,
538
+ "grad_norm": 0.4066360592842102,
539
+ "learning_rate": 7.6e-05,
540
+ "loss": 0.1102,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 2.873134328358209,
545
+ "grad_norm": 0.47184985876083374,
546
+ "learning_rate": 7.7e-05,
547
+ "loss": 0.1219,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 2.91044776119403,
552
+ "grad_norm": 0.6611475348472595,
553
+ "learning_rate": 7.800000000000001e-05,
554
+ "loss": 0.1267,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 2.9477611940298507,
559
+ "grad_norm": 0.3570108413696289,
560
+ "learning_rate": 7.900000000000001e-05,
561
+ "loss": 0.1191,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 2.9850746268656714,
566
+ "grad_norm": 0.4581681489944458,
567
+ "learning_rate": 8e-05,
568
+ "loss": 0.1209,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.0223880597014925,
573
+ "grad_norm": 0.4643435776233673,
574
+ "learning_rate": 8.1e-05,
575
+ "loss": 0.129,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.0597014925373136,
580
+ "grad_norm": 0.5595763921737671,
581
+ "learning_rate": 8.2e-05,
582
+ "loss": 0.1158,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.0970149253731343,
587
+ "grad_norm": 0.48848605155944824,
588
+ "learning_rate": 8.3e-05,
589
+ "loss": 0.1188,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.1343283582089554,
594
+ "grad_norm": 0.4496570825576782,
595
+ "learning_rate": 8.4e-05,
596
+ "loss": 0.114,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.171641791044776,
601
+ "grad_norm": 0.31364986300468445,
602
+ "learning_rate": 8.5e-05,
603
+ "loss": 0.1196,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.208955223880597,
608
+ "grad_norm": 0.3395878076553345,
609
+ "learning_rate": 8.6e-05,
610
+ "loss": 0.1124,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.246268656716418,
615
+ "grad_norm": 0.4917413592338562,
616
+ "learning_rate": 8.7e-05,
617
+ "loss": 0.1074,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.283582089552239,
622
+ "grad_norm": 0.44114553928375244,
623
+ "learning_rate": 8.800000000000001e-05,
624
+ "loss": 0.1095,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.3208955223880596,
629
+ "grad_norm": 0.3323831558227539,
630
+ "learning_rate": 8.900000000000001e-05,
631
+ "loss": 0.106,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.3582089552238807,
636
+ "grad_norm": 0.4495660066604614,
637
+ "learning_rate": 9e-05,
638
+ "loss": 0.1222,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.3955223880597014,
643
+ "grad_norm": 0.40784788131713867,
644
+ "learning_rate": 9.1e-05,
645
+ "loss": 0.1048,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.4328358208955225,
650
+ "grad_norm": 0.4643700420856476,
651
+ "learning_rate": 9.200000000000001e-05,
652
+ "loss": 0.1097,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.470149253731343,
657
+ "grad_norm": 0.472494512796402,
658
+ "learning_rate": 9.300000000000001e-05,
659
+ "loss": 0.1041,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.5074626865671643,
664
+ "grad_norm": 0.6110897660255432,
665
+ "learning_rate": 9.4e-05,
666
+ "loss": 0.0959,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.544776119402985,
671
+ "grad_norm": 0.5313069820404053,
672
+ "learning_rate": 9.5e-05,
673
+ "loss": 0.113,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.582089552238806,
678
+ "grad_norm": 0.4223133623600006,
679
+ "learning_rate": 9.6e-05,
680
+ "loss": 0.099,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.6194029850746268,
685
+ "grad_norm": 0.5464731454849243,
686
+ "learning_rate": 9.7e-05,
687
+ "loss": 0.1008,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 3.656716417910448,
692
+ "grad_norm": 0.3538314402103424,
693
+ "learning_rate": 9.8e-05,
694
+ "loss": 0.1049,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 3.6940298507462686,
699
+ "grad_norm": 0.7460148334503174,
700
+ "learning_rate": 9.900000000000001e-05,
701
+ "loss": 0.1088,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 3.7313432835820897,
706
+ "grad_norm": 0.3210597038269043,
707
+ "learning_rate": 0.0001,
708
+ "loss": 0.1041,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 3.7686567164179103,
713
+ "grad_norm": 0.4450497627258301,
714
+ "learning_rate": 9.999993165095463e-05,
715
+ "loss": 0.0985,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 3.8059701492537314,
720
+ "grad_norm": 0.4348960816860199,
721
+ "learning_rate": 9.999972660400536e-05,
722
+ "loss": 0.1015,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 3.843283582089552,
727
+ "grad_norm": 0.462782621383667,
728
+ "learning_rate": 9.999938485971279e-05,
729
+ "loss": 0.1068,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 3.8805970149253732,
734
+ "grad_norm": 0.3801368474960327,
735
+ "learning_rate": 9.999890641901125e-05,
736
+ "loss": 0.1117,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 3.917910447761194,
741
+ "grad_norm": 0.45135366916656494,
742
+ "learning_rate": 9.999829128320874e-05,
743
+ "loss": 0.0917,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 3.955223880597015,
748
+ "grad_norm": 0.41138389706611633,
749
+ "learning_rate": 9.999753945398704e-05,
750
+ "loss": 0.1049,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 3.9925373134328357,
755
+ "grad_norm": 0.4976252317428589,
756
+ "learning_rate": 9.999665093340165e-05,
757
+ "loss": 0.1029,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.029850746268656,
762
+ "grad_norm": 0.46372008323669434,
763
+ "learning_rate": 9.99956257238817e-05,
764
+ "loss": 0.1012,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.067164179104478,
769
+ "grad_norm": 0.546938955783844,
770
+ "learning_rate": 9.999446382823013e-05,
771
+ "loss": 0.0829,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.104477611940299,
776
+ "grad_norm": 0.40513405203819275,
777
+ "learning_rate": 9.999316524962345e-05,
778
+ "loss": 0.0933,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.141791044776119,
783
+ "grad_norm": 0.4198484420776367,
784
+ "learning_rate": 9.999172999161198e-05,
785
+ "loss": 0.0895,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.17910447761194,
790
+ "grad_norm": 0.3965628743171692,
791
+ "learning_rate": 9.999015805811965e-05,
792
+ "loss": 0.0917,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.2164179104477615,
797
+ "grad_norm": 0.3095884621143341,
798
+ "learning_rate": 9.998844945344405e-05,
799
+ "loss": 0.0953,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.253731343283582,
804
+ "grad_norm": 0.7962276339530945,
805
+ "learning_rate": 9.998660418225645e-05,
806
+ "loss": 0.0979,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.291044776119403,
811
+ "grad_norm": 0.42066490650177,
812
+ "learning_rate": 9.998462224960175e-05,
813
+ "loss": 0.099,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.3283582089552235,
818
+ "grad_norm": 0.3894193470478058,
819
+ "learning_rate": 9.998250366089848e-05,
820
+ "loss": 0.0887,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.365671641791045,
825
+ "grad_norm": 0.28998032212257385,
826
+ "learning_rate": 9.998024842193876e-05,
827
+ "loss": 0.0943,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.402985074626866,
832
+ "grad_norm": 0.3919823467731476,
833
+ "learning_rate": 9.997785653888835e-05,
834
+ "loss": 0.0916,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.440298507462686,
839
+ "grad_norm": 0.3708650469779968,
840
+ "learning_rate": 9.997532801828658e-05,
841
+ "loss": 0.0858,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.477611940298507,
846
+ "grad_norm": 0.2935069799423218,
847
+ "learning_rate": 9.997266286704631e-05,
848
+ "loss": 0.0992,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.514925373134329,
853
+ "grad_norm": 0.4675377607345581,
854
+ "learning_rate": 9.996986109245395e-05,
855
+ "loss": 0.0854,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.552238805970149,
860
+ "grad_norm": 0.31374865770339966,
861
+ "learning_rate": 9.996692270216947e-05,
862
+ "loss": 0.0788,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 4.58955223880597,
867
+ "grad_norm": 0.419249951839447,
868
+ "learning_rate": 9.996384770422629e-05,
869
+ "loss": 0.0873,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 4.6268656716417915,
874
+ "grad_norm": 0.26002731919288635,
875
+ "learning_rate": 9.996063610703137e-05,
876
+ "loss": 0.0845,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 4.664179104477612,
881
+ "grad_norm": 0.29573896527290344,
882
+ "learning_rate": 9.995728791936504e-05,
883
+ "loss": 0.091,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 4.701492537313433,
888
+ "grad_norm": 0.33090147376060486,
889
+ "learning_rate": 9.995380315038119e-05,
890
+ "loss": 0.0827,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 4.7388059701492535,
895
+ "grad_norm": 0.24417485296726227,
896
+ "learning_rate": 9.9950181809607e-05,
897
+ "loss": 0.0859,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 4.776119402985074,
902
+ "grad_norm": 0.48290401697158813,
903
+ "learning_rate": 9.994642390694308e-05,
904
+ "loss": 0.0889,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 4.813432835820896,
909
+ "grad_norm": 0.4479697048664093,
910
+ "learning_rate": 9.99425294526634e-05,
911
+ "loss": 0.097,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 4.850746268656716,
916
+ "grad_norm": 0.3560147285461426,
917
+ "learning_rate": 9.993849845741524e-05,
918
+ "loss": 0.0904,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 4.888059701492537,
923
+ "grad_norm": 0.6645416617393494,
924
+ "learning_rate": 9.99343309322192e-05,
925
+ "loss": 0.0922,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 4.925373134328359,
930
+ "grad_norm": 0.29696759581565857,
931
+ "learning_rate": 9.993002688846913e-05,
932
+ "loss": 0.093,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 4.962686567164179,
937
+ "grad_norm": 0.47146692872047424,
938
+ "learning_rate": 9.992558633793212e-05,
939
+ "loss": 0.085,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.0,
944
+ "grad_norm": 0.3430916368961334,
945
+ "learning_rate": 9.992100929274846e-05,
946
+ "loss": 0.0805,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.037313432835821,
951
+ "grad_norm": 0.3205055892467499,
952
+ "learning_rate": 9.991629576543163e-05,
953
+ "loss": 0.0766,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.074626865671641,
958
+ "grad_norm": 0.3664805293083191,
959
+ "learning_rate": 9.991144576886823e-05,
960
+ "loss": 0.0766,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.111940298507463,
965
+ "grad_norm": 0.3753412663936615,
966
+ "learning_rate": 9.990645931631796e-05,
967
+ "loss": 0.0688,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.149253731343284,
972
+ "grad_norm": 0.31633055210113525,
973
+ "learning_rate": 9.990133642141359e-05,
974
+ "loss": 0.0796,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.186567164179104,
979
+ "grad_norm": 0.3355732262134552,
980
+ "learning_rate": 9.989607709816091e-05,
981
+ "loss": 0.0716,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.223880597014926,
986
+ "grad_norm": 0.24850831925868988,
987
+ "learning_rate": 9.989068136093873e-05,
988
+ "loss": 0.0778,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.2611940298507465,
993
+ "grad_norm": 0.29537102580070496,
994
+ "learning_rate": 9.988514922449879e-05,
995
+ "loss": 0.0759,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.298507462686567,
1000
+ "grad_norm": 0.3430945873260498,
1001
+ "learning_rate": 9.987948070396571e-05,
1002
+ "loss": 0.0774,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.335820895522388,
1007
+ "grad_norm": 0.5220637917518616,
1008
+ "learning_rate": 9.987367581483705e-05,
1009
+ "loss": 0.0836,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.373134328358209,
1014
+ "grad_norm": 0.28184008598327637,
1015
+ "learning_rate": 9.986773457298311e-05,
1016
+ "loss": 0.0752,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.41044776119403,
1021
+ "grad_norm": 0.36261311173439026,
1022
+ "learning_rate": 9.986165699464705e-05,
1023
+ "loss": 0.075,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.447761194029851,
1028
+ "grad_norm": 0.5107380151748657,
1029
+ "learning_rate": 9.985544309644475e-05,
1030
+ "loss": 0.0814,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 5.485074626865671,
1035
+ "grad_norm": 0.2446671426296234,
1036
+ "learning_rate": 9.984909289536473e-05,
1037
+ "loss": 0.0704,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 5.522388059701493,
1042
+ "grad_norm": 0.30449381470680237,
1043
+ "learning_rate": 9.984260640876821e-05,
1044
+ "loss": 0.0794,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 5.559701492537314,
1049
+ "grad_norm": 0.25645050406455994,
1050
+ "learning_rate": 9.983598365438902e-05,
1051
+ "loss": 0.0709,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 5.597014925373134,
1056
+ "grad_norm": 0.23825006186962128,
1057
+ "learning_rate": 9.98292246503335e-05,
1058
+ "loss": 0.0828,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 5.634328358208955,
1063
+ "grad_norm": 0.3259269893169403,
1064
+ "learning_rate": 9.98223294150805e-05,
1065
+ "loss": 0.0824,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 5.6716417910447765,
1070
+ "grad_norm": 0.24058914184570312,
1071
+ "learning_rate": 9.981529796748134e-05,
1072
+ "loss": 0.073,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 5.708955223880597,
1077
+ "grad_norm": 0.34457242488861084,
1078
+ "learning_rate": 9.980813032675974e-05,
1079
+ "loss": 0.0845,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 5.746268656716418,
1084
+ "grad_norm": 0.32940393686294556,
1085
+ "learning_rate": 9.980082651251175e-05,
1086
+ "loss": 0.0832,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 5.7835820895522385,
1091
+ "grad_norm": 0.5683007836341858,
1092
+ "learning_rate": 9.979338654470569e-05,
1093
+ "loss": 0.0836,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 5.82089552238806,
1098
+ "grad_norm": 0.31041061878204346,
1099
+ "learning_rate": 9.97858104436822e-05,
1100
+ "loss": 0.07,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 5.858208955223881,
1105
+ "grad_norm": 0.37858131527900696,
1106
+ "learning_rate": 9.977809823015401e-05,
1107
+ "loss": 0.0738,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 5.895522388059701,
1112
+ "grad_norm": 0.2743091583251953,
1113
+ "learning_rate": 9.977024992520602e-05,
1114
+ "loss": 0.0761,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 5.932835820895522,
1119
+ "grad_norm": 0.29117098450660706,
1120
+ "learning_rate": 9.976226555029522e-05,
1121
+ "loss": 0.0777,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 5.970149253731344,
1126
+ "grad_norm": 0.31398633122444153,
1127
+ "learning_rate": 9.975414512725057e-05,
1128
+ "loss": 0.0664,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.007462686567164,
1133
+ "grad_norm": 0.2684272527694702,
1134
+ "learning_rate": 9.974588867827301e-05,
1135
+ "loss": 0.0686,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.044776119402985,
1140
+ "grad_norm": 0.3945397436618805,
1141
+ "learning_rate": 9.973749622593534e-05,
1142
+ "loss": 0.0614,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.082089552238806,
1147
+ "grad_norm": 0.2747954726219177,
1148
+ "learning_rate": 9.972896779318219e-05,
1149
+ "loss": 0.0681,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.119402985074627,
1154
+ "grad_norm": 0.43257200717926025,
1155
+ "learning_rate": 9.972030340333001e-05,
1156
+ "loss": 0.0725,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.156716417910448,
1161
+ "grad_norm": 0.3559250831604004,
1162
+ "learning_rate": 9.97115030800669e-05,
1163
+ "loss": 0.0804,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.1940298507462686,
1168
+ "grad_norm": 0.3079264760017395,
1169
+ "learning_rate": 9.970256684745258e-05,
1170
+ "loss": 0.0649,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.231343283582089,
1175
+ "grad_norm": 0.32298946380615234,
1176
+ "learning_rate": 9.969349472991838e-05,
1177
+ "loss": 0.0668,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.268656716417911,
1182
+ "grad_norm": 0.2826225459575653,
1183
+ "learning_rate": 9.968428675226714e-05,
1184
+ "loss": 0.0734,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.3059701492537314,
1189
+ "grad_norm": 0.39002349972724915,
1190
+ "learning_rate": 9.967494293967312e-05,
1191
+ "loss": 0.0728,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.343283582089552,
1196
+ "grad_norm": 0.403890997171402,
1197
+ "learning_rate": 9.966546331768191e-05,
1198
+ "loss": 0.067,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.380597014925373,
1203
+ "grad_norm": 0.3755359351634979,
1204
+ "learning_rate": 9.965584791221048e-05,
1205
+ "loss": 0.0755,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 6.417910447761194,
1210
+ "grad_norm": 0.26346635818481445,
1211
+ "learning_rate": 9.964609674954696e-05,
1212
+ "loss": 0.0728,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 6.455223880597015,
1217
+ "grad_norm": 0.45292145013809204,
1218
+ "learning_rate": 9.963620985635065e-05,
1219
+ "loss": 0.0731,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 6.492537313432836,
1224
+ "grad_norm": 0.3568434715270996,
1225
+ "learning_rate": 9.962618725965196e-05,
1226
+ "loss": 0.0761,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 6.529850746268656,
1231
+ "grad_norm": 0.2551257014274597,
1232
+ "learning_rate": 9.961602898685226e-05,
1233
+ "loss": 0.0694,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 6.567164179104478,
1238
+ "grad_norm": 0.6106354594230652,
1239
+ "learning_rate": 9.96057350657239e-05,
1240
+ "loss": 0.0827,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 6.604477611940299,
1245
+ "grad_norm": 0.3226093053817749,
1246
+ "learning_rate": 9.959530552441005e-05,
1247
+ "loss": 0.0716,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 6.641791044776119,
1252
+ "grad_norm": 0.4297254979610443,
1253
+ "learning_rate": 9.95847403914247e-05,
1254
+ "loss": 0.0748,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 6.67910447761194,
1259
+ "grad_norm": 0.26469680666923523,
1260
+ "learning_rate": 9.95740396956525e-05,
1261
+ "loss": 0.074,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 6.7164179104477615,
1266
+ "grad_norm": 0.22717897593975067,
1267
+ "learning_rate": 9.956320346634876e-05,
1268
+ "loss": 0.0739,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 6.753731343283582,
1273
+ "grad_norm": 0.4513498544692993,
1274
+ "learning_rate": 9.955223173313931e-05,
1275
+ "loss": 0.0664,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 6.791044776119403,
1280
+ "grad_norm": 0.31683439016342163,
1281
+ "learning_rate": 9.954112452602045e-05,
1282
+ "loss": 0.069,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 6.8283582089552235,
1287
+ "grad_norm": 0.3350532650947571,
1288
+ "learning_rate": 9.952988187535886e-05,
1289
+ "loss": 0.0699,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 6.865671641791045,
1294
+ "grad_norm": 0.29829463362693787,
1295
+ "learning_rate": 9.95185038118915e-05,
1296
+ "loss": 0.0663,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 6.902985074626866,
1301
+ "grad_norm": 0.31650781631469727,
1302
+ "learning_rate": 9.950699036672559e-05,
1303
+ "loss": 0.0668,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 6.940298507462686,
1308
+ "grad_norm": 0.360944926738739,
1309
+ "learning_rate": 9.949534157133844e-05,
1310
+ "loss": 0.0696,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 6.977611940298507,
1315
+ "grad_norm": 0.31337013840675354,
1316
+ "learning_rate": 9.948355745757741e-05,
1317
+ "loss": 0.073,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.014925373134329,
1322
+ "grad_norm": 0.4675919711589813,
1323
+ "learning_rate": 9.94716380576598e-05,
1324
+ "loss": 0.0688,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.052238805970149,
1329
+ "grad_norm": 0.3031919002532959,
1330
+ "learning_rate": 9.945958340417283e-05,
1331
+ "loss": 0.0596,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.08955223880597,
1336
+ "grad_norm": 0.24858474731445312,
1337
+ "learning_rate": 9.944739353007344e-05,
1338
+ "loss": 0.0717,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.126865671641791,
1343
+ "grad_norm": 0.20959483087062836,
1344
+ "learning_rate": 9.943506846868826e-05,
1345
+ "loss": 0.0694,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.164179104477612,
1350
+ "grad_norm": 0.35621434450149536,
1351
+ "learning_rate": 9.942260825371358e-05,
1352
+ "loss": 0.063,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.201492537313433,
1357
+ "grad_norm": 0.3462587594985962,
1358
+ "learning_rate": 9.941001291921512e-05,
1359
+ "loss": 0.068,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.2388059701492535,
1364
+ "grad_norm": 0.38649681210517883,
1365
+ "learning_rate": 9.939728249962807e-05,
1366
+ "loss": 0.0638,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.276119402985074,
1371
+ "grad_norm": 0.29564595222473145,
1372
+ "learning_rate": 9.938441702975689e-05,
1373
+ "loss": 0.0626,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 7.313432835820896,
1378
+ "grad_norm": 0.339857816696167,
1379
+ "learning_rate": 9.937141654477528e-05,
1380
+ "loss": 0.0535,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 7.350746268656716,
1385
+ "grad_norm": 0.2591215670108795,
1386
+ "learning_rate": 9.93582810802261e-05,
1387
+ "loss": 0.0645,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 7.388059701492537,
1392
+ "grad_norm": 0.30237796902656555,
1393
+ "learning_rate": 9.934501067202117e-05,
1394
+ "loss": 0.0675,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 7.425373134328359,
1399
+ "grad_norm": 0.28394174575805664,
1400
+ "learning_rate": 9.93316053564413e-05,
1401
+ "loss": 0.0643,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 7.462686567164179,
1406
+ "grad_norm": 0.3124663233757019,
1407
+ "learning_rate": 9.931806517013612e-05,
1408
+ "loss": 0.059,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 7.5,
1413
+ "grad_norm": 0.36073037981987,
1414
+ "learning_rate": 9.930439015012396e-05,
1415
+ "loss": 0.0606,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 7.537313432835821,
1420
+ "grad_norm": 0.4091481864452362,
1421
+ "learning_rate": 9.929058033379181e-05,
1422
+ "loss": 0.0603,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 7.574626865671641,
1427
+ "grad_norm": 0.44718074798583984,
1428
+ "learning_rate": 9.927663575889521e-05,
1429
+ "loss": 0.0741,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 7.611940298507463,
1434
+ "grad_norm": 0.3819601833820343,
1435
+ "learning_rate": 9.926255646355804e-05,
1436
+ "loss": 0.0707,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 7.649253731343284,
1441
+ "grad_norm": 0.23336420953273773,
1442
+ "learning_rate": 9.92483424862726e-05,
1443
+ "loss": 0.0676,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 7.686567164179104,
1448
+ "grad_norm": 0.24415315687656403,
1449
+ "learning_rate": 9.923399386589933e-05,
1450
+ "loss": 0.0594,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 7.723880597014926,
1455
+ "grad_norm": 0.3735473155975342,
1456
+ "learning_rate": 9.921951064166684e-05,
1457
+ "loss": 0.062,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 7.7611940298507465,
1462
+ "grad_norm": 0.31629472970962524,
1463
+ "learning_rate": 9.92048928531717e-05,
1464
+ "loss": 0.0606,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 7.798507462686567,
1469
+ "grad_norm": 0.37902557849884033,
1470
+ "learning_rate": 9.919014054037836e-05,
1471
+ "loss": 0.0584,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 7.835820895522388,
1476
+ "grad_norm": 0.3486720323562622,
1477
+ "learning_rate": 9.917525374361912e-05,
1478
+ "loss": 0.056,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 7.8731343283582085,
1483
+ "grad_norm": 0.3731362521648407,
1484
+ "learning_rate": 9.91602325035939e-05,
1485
+ "loss": 0.0601,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 7.91044776119403,
1490
+ "grad_norm": 0.3560399115085602,
1491
+ "learning_rate": 9.914507686137019e-05,
1492
+ "loss": 0.06,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 7.947761194029851,
1497
+ "grad_norm": 0.30075564980506897,
1498
+ "learning_rate": 9.912978685838294e-05,
1499
+ "loss": 0.0657,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 7.985074626865671,
1504
+ "grad_norm": 0.2984028458595276,
1505
+ "learning_rate": 9.911436253643445e-05,
1506
+ "loss": 0.0587,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 8.022388059701493,
1511
+ "grad_norm": 0.1980169117450714,
1512
+ "learning_rate": 9.90988039376942e-05,
1513
+ "loss": 0.0718,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 8.059701492537313,
1518
+ "grad_norm": 0.31339579820632935,
1519
+ "learning_rate": 9.90831111046988e-05,
1520
+ "loss": 0.0557,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 8.097014925373134,
1525
+ "grad_norm": 0.1968696266412735,
1526
+ "learning_rate": 9.90672840803519e-05,
1527
+ "loss": 0.0571,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 8.134328358208956,
1532
+ "grad_norm": 0.23931682109832764,
1533
+ "learning_rate": 9.905132290792394e-05,
1534
+ "loss": 0.0566,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 8.171641791044776,
1539
+ "grad_norm": 0.21741189062595367,
1540
+ "learning_rate": 9.903522763105218e-05,
1541
+ "loss": 0.0575,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 8.208955223880597,
1546
+ "grad_norm": 0.22874368727207184,
1547
+ "learning_rate": 9.901899829374047e-05,
1548
+ "loss": 0.0565,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 8.246268656716419,
1553
+ "grad_norm": 0.3441888093948364,
1554
+ "learning_rate": 9.900263494035921e-05,
1555
+ "loss": 0.0565,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 8.283582089552239,
1560
+ "grad_norm": 0.2539830803871155,
1561
+ "learning_rate": 9.89861376156452e-05,
1562
+ "loss": 0.0538,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 8.32089552238806,
1567
+ "grad_norm": 0.2235102653503418,
1568
+ "learning_rate": 9.896950636470147e-05,
1569
+ "loss": 0.0609,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 8.35820895522388,
1574
+ "grad_norm": 0.1941322684288025,
1575
+ "learning_rate": 9.895274123299723e-05,
1576
+ "loss": 0.0562,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 8.395522388059701,
1581
+ "grad_norm": 0.2691369950771332,
1582
+ "learning_rate": 9.893584226636772e-05,
1583
+ "loss": 0.0608,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 8.432835820895523,
1588
+ "grad_norm": 0.24730461835861206,
1589
+ "learning_rate": 9.891880951101407e-05,
1590
+ "loss": 0.0582,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 8.470149253731343,
1595
+ "grad_norm": 0.34785839915275574,
1596
+ "learning_rate": 9.890164301350318e-05,
1597
+ "loss": 0.0506,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 8.507462686567164,
1602
+ "grad_norm": 0.3625825345516205,
1603
+ "learning_rate": 9.888434282076758e-05,
1604
+ "loss": 0.0614,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 8.544776119402986,
1609
+ "grad_norm": 0.25210148096084595,
1610
+ "learning_rate": 9.886690898010535e-05,
1611
+ "loss": 0.0611,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 8.582089552238806,
1616
+ "grad_norm": 0.27312466502189636,
1617
+ "learning_rate": 9.884934153917997e-05,
1618
+ "loss": 0.0537,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 8.619402985074627,
1623
+ "grad_norm": 0.314647912979126,
1624
+ "learning_rate": 9.883164054602012e-05,
1625
+ "loss": 0.0602,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 8.656716417910447,
1630
+ "grad_norm": 0.21531912684440613,
1631
+ "learning_rate": 9.881380604901964e-05,
1632
+ "loss": 0.0552,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 8.694029850746269,
1637
+ "grad_norm": 0.23920664191246033,
1638
+ "learning_rate": 9.879583809693738e-05,
1639
+ "loss": 0.0613,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 8.73134328358209,
1644
+ "grad_norm": 0.21864956617355347,
1645
+ "learning_rate": 9.877773673889701e-05,
1646
+ "loss": 0.0649,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 8.76865671641791,
1651
+ "grad_norm": 0.27523377537727356,
1652
+ "learning_rate": 9.8759502024387e-05,
1653
+ "loss": 0.0606,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 8.805970149253731,
1658
+ "grad_norm": 0.24805469810962677,
1659
+ "learning_rate": 9.87411340032603e-05,
1660
+ "loss": 0.0549,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 8.843283582089553,
1665
+ "grad_norm": 0.23070092499256134,
1666
+ "learning_rate": 9.872263272573443e-05,
1667
+ "loss": 0.0562,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 8.880597014925373,
1672
+ "grad_norm": 0.20833946764469147,
1673
+ "learning_rate": 9.870399824239117e-05,
1674
+ "loss": 0.05,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 8.917910447761194,
1679
+ "grad_norm": 0.34507372975349426,
1680
+ "learning_rate": 9.868523060417646e-05,
1681
+ "loss": 0.0613,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 8.955223880597014,
1686
+ "grad_norm": 0.32865110039711,
1687
+ "learning_rate": 9.86663298624003e-05,
1688
+ "loss": 0.0621,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 8.992537313432836,
1693
+ "grad_norm": 0.21305270493030548,
1694
+ "learning_rate": 9.864729606873663e-05,
1695
+ "loss": 0.0572,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 9.029850746268657,
1700
+ "grad_norm": 0.28193730115890503,
1701
+ "learning_rate": 9.862812927522309e-05,
1702
+ "loss": 0.0555,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 9.067164179104477,
1707
+ "grad_norm": 0.3953789472579956,
1708
+ "learning_rate": 9.860882953426099e-05,
1709
+ "loss": 0.0536,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 9.104477611940299,
1714
+ "grad_norm": 0.23013322055339813,
1715
+ "learning_rate": 9.858939689861506e-05,
1716
+ "loss": 0.0572,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 9.14179104477612,
1721
+ "grad_norm": 0.2906680107116699,
1722
+ "learning_rate": 9.856983142141339e-05,
1723
+ "loss": 0.0592,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 9.17910447761194,
1728
+ "grad_norm": 0.23490828275680542,
1729
+ "learning_rate": 9.855013315614725e-05,
1730
+ "loss": 0.0583,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 9.216417910447761,
1735
+ "grad_norm": 0.22825880348682404,
1736
+ "learning_rate": 9.853030215667093e-05,
1737
+ "loss": 0.059,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 9.253731343283581,
1742
+ "grad_norm": 0.25871285796165466,
1743
+ "learning_rate": 9.851033847720166e-05,
1744
+ "loss": 0.0555,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 9.291044776119403,
1749
+ "grad_norm": 0.27220776677131653,
1750
+ "learning_rate": 9.849024217231935e-05,
1751
+ "loss": 0.0542,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 9.328358208955224,
1756
+ "grad_norm": 0.26534005999565125,
1757
+ "learning_rate": 9.847001329696653e-05,
1758
+ "loss": 0.0526,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 9.365671641791044,
1763
+ "grad_norm": 0.33486032485961914,
1764
+ "learning_rate": 9.844965190644817e-05,
1765
+ "loss": 0.0563,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 9.402985074626866,
1770
+ "grad_norm": 0.2949483394622803,
1771
+ "learning_rate": 9.842915805643155e-05,
1772
+ "loss": 0.0556,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 9.440298507462687,
1777
+ "grad_norm": 0.24123981595039368,
1778
+ "learning_rate": 9.840853180294608e-05,
1779
+ "loss": 0.05,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 9.477611940298507,
1784
+ "grad_norm": 0.22536049783229828,
1785
+ "learning_rate": 9.838777320238312e-05,
1786
+ "loss": 0.0522,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 9.514925373134329,
1791
+ "grad_norm": 0.23206663131713867,
1792
+ "learning_rate": 9.836688231149592e-05,
1793
+ "loss": 0.0591,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 9.552238805970148,
1798
+ "grad_norm": 0.28573134541511536,
1799
+ "learning_rate": 9.834585918739936e-05,
1800
+ "loss": 0.0568,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 9.58955223880597,
1805
+ "grad_norm": 0.2628820538520813,
1806
+ "learning_rate": 9.832470388756987e-05,
1807
+ "loss": 0.0571,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 9.626865671641792,
1812
+ "grad_norm": 0.2880440652370453,
1813
+ "learning_rate": 9.830341646984521e-05,
1814
+ "loss": 0.0559,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 9.664179104477611,
1819
+ "grad_norm": 0.1786259263753891,
1820
+ "learning_rate": 9.82819969924244e-05,
1821
+ "loss": 0.058,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 9.701492537313433,
1826
+ "grad_norm": 0.3501608073711395,
1827
+ "learning_rate": 9.826044551386744e-05,
1828
+ "loss": 0.0523,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 9.738805970149254,
1833
+ "grad_norm": 0.24757252633571625,
1834
+ "learning_rate": 9.823876209309527e-05,
1835
+ "loss": 0.0587,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 9.776119402985074,
1840
+ "grad_norm": 0.2556290626525879,
1841
+ "learning_rate": 9.821694678938953e-05,
1842
+ "loss": 0.0555,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 9.813432835820896,
1847
+ "grad_norm": 0.2561217248439789,
1848
+ "learning_rate": 9.819499966239243e-05,
1849
+ "loss": 0.052,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 9.850746268656717,
1854
+ "grad_norm": 0.2776634097099304,
1855
+ "learning_rate": 9.817292077210659e-05,
1856
+ "loss": 0.0498,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 9.888059701492537,
1861
+ "grad_norm": 0.20668549835681915,
1862
+ "learning_rate": 9.815071017889482e-05,
1863
+ "loss": 0.0517,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 9.925373134328359,
1868
+ "grad_norm": 0.3100263178348541,
1869
+ "learning_rate": 9.812836794348004e-05,
1870
+ "loss": 0.0633,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 9.962686567164178,
1875
+ "grad_norm": 0.2780782878398895,
1876
+ "learning_rate": 9.81058941269451e-05,
1877
+ "loss": 0.0581,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 10.0,
1882
+ "grad_norm": 0.28903728723526,
1883
+ "learning_rate": 9.808328879073251e-05,
1884
+ "loss": 0.0538,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 10.037313432835822,
1889
+ "grad_norm": 0.22727562487125397,
1890
+ "learning_rate": 9.806055199664446e-05,
1891
+ "loss": 0.0491,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 10.074626865671641,
1896
+ "grad_norm": 0.267918199300766,
1897
+ "learning_rate": 9.803768380684242e-05,
1898
+ "loss": 0.0562,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 10.111940298507463,
1903
+ "grad_norm": 0.2988606095314026,
1904
+ "learning_rate": 9.801468428384716e-05,
1905
+ "loss": 0.0566,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 10.149253731343283,
1910
+ "grad_norm": 0.2710281312465668,
1911
+ "learning_rate": 9.799155349053851e-05,
1912
+ "loss": 0.0541,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 10.186567164179104,
1917
+ "grad_norm": 0.15320520102977753,
1918
+ "learning_rate": 9.796829149015517e-05,
1919
+ "loss": 0.0548,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 10.223880597014926,
1924
+ "grad_norm": 0.2653089463710785,
1925
+ "learning_rate": 9.794489834629455e-05,
1926
+ "loss": 0.0599,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 10.261194029850746,
1931
+ "grad_norm": 0.19223959743976593,
1932
+ "learning_rate": 9.792137412291265e-05,
1933
+ "loss": 0.0494,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 10.298507462686567,
1938
+ "grad_norm": 0.20455987751483917,
1939
+ "learning_rate": 9.789771888432375e-05,
1940
+ "loss": 0.0538,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 10.335820895522389,
1945
+ "grad_norm": 0.24908749759197235,
1946
+ "learning_rate": 9.787393269520039e-05,
1947
+ "loss": 0.0481,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 10.373134328358208,
1952
+ "grad_norm": 0.3131813406944275,
1953
+ "learning_rate": 9.785001562057309e-05,
1954
+ "loss": 0.0526,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 10.41044776119403,
1959
+ "grad_norm": 0.24828971922397614,
1960
+ "learning_rate": 9.782596772583026e-05,
1961
+ "loss": 0.0489,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 10.447761194029852,
1966
+ "grad_norm": 0.21727119386196136,
1967
+ "learning_rate": 9.780178907671789e-05,
1968
+ "loss": 0.0532,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 10.485074626865671,
1973
+ "grad_norm": 0.20279547572135925,
1974
+ "learning_rate": 9.777747973933948e-05,
1975
+ "loss": 0.0565,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 10.522388059701493,
1980
+ "grad_norm": 0.17726702988147736,
1981
+ "learning_rate": 9.775303978015585e-05,
1982
+ "loss": 0.0437,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 10.559701492537313,
1987
+ "grad_norm": 0.18961119651794434,
1988
+ "learning_rate": 9.772846926598491e-05,
1989
+ "loss": 0.0584,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 10.597014925373134,
1994
+ "grad_norm": 0.2498980015516281,
1995
+ "learning_rate": 9.77037682640015e-05,
1996
+ "loss": 0.0496,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 10.634328358208956,
2001
+ "grad_norm": 0.16978798806667328,
2002
+ "learning_rate": 9.767893684173721e-05,
2003
+ "loss": 0.0469,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 10.671641791044776,
2008
+ "grad_norm": 0.16128584742546082,
2009
+ "learning_rate": 9.765397506708023e-05,
2010
+ "loss": 0.0533,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 10.708955223880597,
2015
+ "grad_norm": 0.20463155210018158,
2016
+ "learning_rate": 9.762888300827507e-05,
2017
+ "loss": 0.0464,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 10.746268656716419,
2022
+ "grad_norm": 0.30601629614830017,
2023
+ "learning_rate": 9.760366073392246e-05,
2024
+ "loss": 0.0489,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 10.783582089552239,
2029
+ "grad_norm": 0.2730671763420105,
2030
+ "learning_rate": 9.757830831297914e-05,
2031
+ "loss": 0.0495,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 10.82089552238806,
2036
+ "grad_norm": 0.251432865858078,
2037
+ "learning_rate": 9.755282581475769e-05,
2038
+ "loss": 0.0549,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 10.85820895522388,
2043
+ "grad_norm": 0.26670166850090027,
2044
+ "learning_rate": 9.752721330892624e-05,
2045
+ "loss": 0.061,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 10.895522388059701,
2050
+ "grad_norm": 0.2965967655181885,
2051
+ "learning_rate": 9.750147086550844e-05,
2052
+ "loss": 0.0473,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 10.932835820895523,
2057
+ "grad_norm": 0.683840274810791,
2058
+ "learning_rate": 9.747559855488313e-05,
2059
+ "loss": 0.0509,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 10.970149253731343,
2064
+ "grad_norm": 0.25740495324134827,
2065
+ "learning_rate": 9.744959644778422e-05,
2066
+ "loss": 0.0515,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 11.007462686567164,
2071
+ "grad_norm": 0.2880542278289795,
2072
+ "learning_rate": 9.742346461530048e-05,
2073
+ "loss": 0.0482,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 11.044776119402986,
2078
+ "grad_norm": 0.45032551884651184,
2079
+ "learning_rate": 9.739720312887535e-05,
2080
+ "loss": 0.0557,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 11.082089552238806,
2085
+ "grad_norm": 0.2829900085926056,
2086
+ "learning_rate": 9.73708120603067e-05,
2087
+ "loss": 0.052,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 11.119402985074627,
2092
+ "grad_norm": 0.309597373008728,
2093
+ "learning_rate": 9.734429148174675e-05,
2094
+ "loss": 0.0541,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 11.156716417910447,
2099
+ "grad_norm": 0.2433389127254486,
2100
+ "learning_rate": 9.731764146570173e-05,
2101
+ "loss": 0.0482,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 11.194029850746269,
2106
+ "grad_norm": 0.24458132684230804,
2107
+ "learning_rate": 9.729086208503174e-05,
2108
+ "loss": 0.0505,
2109
+ "step": 3000
2110
+ },
2111
+ {
2112
+ "epoch": 11.23134328358209,
2113
+ "grad_norm": 0.2305087298154831,
2114
+ "learning_rate": 9.726395341295062e-05,
2115
+ "loss": 0.0504,
2116
+ "step": 3010
2117
+ },
2118
+ {
2119
+ "epoch": 11.26865671641791,
2120
+ "grad_norm": 0.18110457062721252,
2121
+ "learning_rate": 9.723691552302562e-05,
2122
+ "loss": 0.0575,
2123
+ "step": 3020
2124
+ },
2125
+ {
2126
+ "epoch": 11.305970149253731,
2127
+ "grad_norm": 0.20407621562480927,
2128
+ "learning_rate": 9.720974848917735e-05,
2129
+ "loss": 0.0494,
2130
+ "step": 3030
2131
+ },
2132
+ {
2133
+ "epoch": 11.343283582089553,
2134
+ "grad_norm": 0.25924697518348694,
2135
+ "learning_rate": 9.718245238567939e-05,
2136
+ "loss": 0.0472,
2137
+ "step": 3040
2138
+ },
2139
+ {
2140
+ "epoch": 11.380597014925373,
2141
+ "grad_norm": 0.23041822016239166,
2142
+ "learning_rate": 9.715502728715826e-05,
2143
+ "loss": 0.0481,
2144
+ "step": 3050
2145
+ },
2146
+ {
2147
+ "epoch": 11.417910447761194,
2148
+ "grad_norm": 0.25381171703338623,
2149
+ "learning_rate": 9.712747326859315e-05,
2150
+ "loss": 0.0543,
2151
+ "step": 3060
2152
+ },
2153
+ {
2154
+ "epoch": 11.455223880597014,
2155
+ "grad_norm": 0.18027640879154205,
2156
+ "learning_rate": 9.709979040531569e-05,
2157
+ "loss": 0.055,
2158
+ "step": 3070
2159
+ },
2160
+ {
2161
+ "epoch": 11.492537313432836,
2162
+ "grad_norm": 0.2954868674278259,
2163
+ "learning_rate": 9.707197877300974e-05,
2164
+ "loss": 0.0473,
2165
+ "step": 3080
2166
+ },
2167
+ {
2168
+ "epoch": 11.529850746268657,
2169
+ "grad_norm": 0.25323861837387085,
2170
+ "learning_rate": 9.704403844771128e-05,
2171
+ "loss": 0.0509,
2172
+ "step": 3090
2173
+ },
2174
+ {
2175
+ "epoch": 11.567164179104477,
2176
+ "grad_norm": 0.36910176277160645,
2177
+ "learning_rate": 9.701596950580806e-05,
2178
+ "loss": 0.0504,
2179
+ "step": 3100
2180
+ },
2181
+ {
2182
+ "epoch": 11.604477611940299,
2183
+ "grad_norm": 0.34199246764183044,
2184
+ "learning_rate": 9.698777202403953e-05,
2185
+ "loss": 0.0526,
2186
+ "step": 3110
2187
+ },
2188
+ {
2189
+ "epoch": 11.64179104477612,
2190
+ "grad_norm": 0.2146557718515396,
2191
+ "learning_rate": 9.695944607949649e-05,
2192
+ "loss": 0.0579,
2193
+ "step": 3120
2194
+ },
2195
+ {
2196
+ "epoch": 11.67910447761194,
2197
+ "grad_norm": 0.20559175312519073,
2198
+ "learning_rate": 9.693099174962103e-05,
2199
+ "loss": 0.0514,
2200
+ "step": 3130
2201
+ },
2202
+ {
2203
+ "epoch": 11.716417910447761,
2204
+ "grad_norm": 0.2689419090747833,
2205
+ "learning_rate": 9.690240911220618e-05,
2206
+ "loss": 0.0534,
2207
+ "step": 3140
2208
+ },
2209
+ {
2210
+ "epoch": 11.753731343283581,
2211
+ "grad_norm": 0.34870603680610657,
2212
+ "learning_rate": 9.687369824539577e-05,
2213
+ "loss": 0.0485,
2214
+ "step": 3150
2215
+ },
2216
+ {
2217
+ "epoch": 11.791044776119403,
2218
+ "grad_norm": 0.15433363616466522,
2219
+ "learning_rate": 9.684485922768422e-05,
2220
+ "loss": 0.0418,
2221
+ "step": 3160
2222
+ },
2223
+ {
2224
+ "epoch": 11.828358208955224,
2225
+ "grad_norm": 0.26874423027038574,
2226
+ "learning_rate": 9.681589213791633e-05,
2227
+ "loss": 0.0537,
2228
+ "step": 3170
2229
+ },
2230
+ {
2231
+ "epoch": 11.865671641791044,
2232
+ "grad_norm": 0.3361654281616211,
2233
+ "learning_rate": 9.6786797055287e-05,
2234
+ "loss": 0.0474,
2235
+ "step": 3180
2236
+ },
2237
+ {
2238
+ "epoch": 11.902985074626866,
2239
+ "grad_norm": 0.17938771843910217,
2240
+ "learning_rate": 9.675757405934103e-05,
2241
+ "loss": 0.0443,
2242
+ "step": 3190
2243
+ },
2244
+ {
2245
+ "epoch": 11.940298507462687,
2246
+ "grad_norm": 0.31368622183799744,
2247
+ "learning_rate": 9.672822322997305e-05,
2248
+ "loss": 0.0594,
2249
+ "step": 3200
2250
+ },
2251
+ {
2252
+ "epoch": 11.977611940298507,
2253
+ "grad_norm": 0.16268151998519897,
2254
+ "learning_rate": 9.669874464742705e-05,
2255
+ "loss": 0.0487,
2256
+ "step": 3210
2257
+ },
2258
+ {
2259
+ "epoch": 12.014925373134329,
2260
+ "grad_norm": 0.23879969120025635,
2261
+ "learning_rate": 9.66691383922964e-05,
2262
+ "loss": 0.0484,
2263
+ "step": 3220
2264
+ },
2265
+ {
2266
+ "epoch": 12.052238805970148,
2267
+ "grad_norm": 0.2321789413690567,
2268
+ "learning_rate": 9.663940454552342e-05,
2269
+ "loss": 0.051,
2270
+ "step": 3230
2271
+ },
2272
+ {
2273
+ "epoch": 12.08955223880597,
2274
+ "grad_norm": 0.22873088717460632,
2275
+ "learning_rate": 9.660954318839933e-05,
2276
+ "loss": 0.0406,
2277
+ "step": 3240
2278
+ },
2279
+ {
2280
+ "epoch": 12.126865671641792,
2281
+ "grad_norm": 0.3767557740211487,
2282
+ "learning_rate": 9.657955440256395e-05,
2283
+ "loss": 0.0432,
2284
+ "step": 3250
2285
+ },
2286
+ {
2287
+ "epoch": 12.164179104477611,
2288
+ "grad_norm": 0.21569453179836273,
2289
+ "learning_rate": 9.654943827000548e-05,
2290
+ "loss": 0.0528,
2291
+ "step": 3260
2292
+ },
2293
+ {
2294
+ "epoch": 12.201492537313433,
2295
+ "grad_norm": 0.23698291182518005,
2296
+ "learning_rate": 9.651919487306025e-05,
2297
+ "loss": 0.0457,
2298
+ "step": 3270
2299
+ },
2300
+ {
2301
+ "epoch": 12.238805970149254,
2302
+ "grad_norm": 0.21086478233337402,
2303
+ "learning_rate": 9.648882429441257e-05,
2304
+ "loss": 0.0508,
2305
+ "step": 3280
2306
+ },
2307
+ {
2308
+ "epoch": 12.276119402985074,
2309
+ "grad_norm": 0.19763463735580444,
2310
+ "learning_rate": 9.645832661709444e-05,
2311
+ "loss": 0.0497,
2312
+ "step": 3290
2313
+ },
2314
+ {
2315
+ "epoch": 12.313432835820896,
2316
+ "grad_norm": 0.18413852155208588,
2317
+ "learning_rate": 9.642770192448536e-05,
2318
+ "loss": 0.0441,
2319
+ "step": 3300
2320
+ },
2321
+ {
2322
+ "epoch": 12.350746268656717,
2323
+ "grad_norm": 0.13946911692619324,
2324
+ "learning_rate": 9.639695030031204e-05,
2325
+ "loss": 0.0453,
2326
+ "step": 3310
2327
+ },
2328
+ {
2329
+ "epoch": 12.388059701492537,
2330
+ "grad_norm": 0.21613670885562897,
2331
+ "learning_rate": 9.636607182864827e-05,
2332
+ "loss": 0.0511,
2333
+ "step": 3320
2334
+ },
2335
+ {
2336
+ "epoch": 12.425373134328359,
2337
+ "grad_norm": 0.24953646957874298,
2338
+ "learning_rate": 9.63350665939146e-05,
2339
+ "loss": 0.0451,
2340
+ "step": 3330
2341
+ },
2342
+ {
2343
+ "epoch": 12.462686567164178,
2344
+ "grad_norm": 0.2993795871734619,
2345
+ "learning_rate": 9.630393468087818e-05,
2346
+ "loss": 0.0469,
2347
+ "step": 3340
2348
+ },
2349
+ {
2350
+ "epoch": 12.5,
2351
+ "grad_norm": 0.2261819839477539,
2352
+ "learning_rate": 9.627267617465243e-05,
2353
+ "loss": 0.0484,
2354
+ "step": 3350
2355
+ },
2356
+ {
2357
+ "epoch": 12.537313432835822,
2358
+ "grad_norm": 0.23026186227798462,
2359
+ "learning_rate": 9.624129116069694e-05,
2360
+ "loss": 0.0452,
2361
+ "step": 3360
2362
+ },
2363
+ {
2364
+ "epoch": 12.574626865671641,
2365
+ "grad_norm": 0.27859947085380554,
2366
+ "learning_rate": 9.620977972481716e-05,
2367
+ "loss": 0.0593,
2368
+ "step": 3370
2369
+ },
2370
+ {
2371
+ "epoch": 12.611940298507463,
2372
+ "grad_norm": 0.23060785233974457,
2373
+ "learning_rate": 9.617814195316411e-05,
2374
+ "loss": 0.05,
2375
+ "step": 3380
2376
+ },
2377
+ {
2378
+ "epoch": 12.649253731343283,
2379
+ "grad_norm": 0.20185025036334991,
2380
+ "learning_rate": 9.614637793223425e-05,
2381
+ "loss": 0.0573,
2382
+ "step": 3390
2383
+ },
2384
+ {
2385
+ "epoch": 12.686567164179104,
2386
+ "grad_norm": 0.3584498167037964,
2387
+ "learning_rate": 9.611448774886924e-05,
2388
+ "loss": 0.052,
2389
+ "step": 3400
2390
+ },
2391
+ {
2392
+ "epoch": 12.723880597014926,
2393
+ "grad_norm": 0.19336827099323273,
2394
+ "learning_rate": 9.60824714902556e-05,
2395
+ "loss": 0.0535,
2396
+ "step": 3410
2397
+ },
2398
+ {
2399
+ "epoch": 12.761194029850746,
2400
+ "grad_norm": 0.22223635017871857,
2401
+ "learning_rate": 9.605032924392457e-05,
2402
+ "loss": 0.05,
2403
+ "step": 3420
2404
+ },
2405
+ {
2406
+ "epoch": 12.798507462686567,
2407
+ "grad_norm": 0.17108851671218872,
2408
+ "learning_rate": 9.601806109775179e-05,
2409
+ "loss": 0.0475,
2410
+ "step": 3430
2411
+ },
2412
+ {
2413
+ "epoch": 12.835820895522389,
2414
+ "grad_norm": 0.3861902952194214,
2415
+ "learning_rate": 9.598566713995718e-05,
2416
+ "loss": 0.0439,
2417
+ "step": 3440
2418
+ },
2419
+ {
2420
+ "epoch": 12.873134328358208,
2421
+ "grad_norm": 0.18927253782749176,
2422
+ "learning_rate": 9.595314745910456e-05,
2423
+ "loss": 0.052,
2424
+ "step": 3450
2425
+ },
2426
+ {
2427
+ "epoch": 12.91044776119403,
2428
+ "grad_norm": 0.21963383257389069,
2429
+ "learning_rate": 9.59205021441015e-05,
2430
+ "loss": 0.0504,
2431
+ "step": 3460
2432
+ },
2433
+ {
2434
+ "epoch": 12.947761194029852,
2435
+ "grad_norm": 0.18016670644283295,
2436
+ "learning_rate": 9.588773128419906e-05,
2437
+ "loss": 0.0467,
2438
+ "step": 3470
2439
+ },
2440
+ {
2441
+ "epoch": 12.985074626865671,
2442
+ "grad_norm": 0.1776365041732788,
2443
+ "learning_rate": 9.58548349689915e-05,
2444
+ "loss": 0.0414,
2445
+ "step": 3480
2446
+ },
2447
+ {
2448
+ "epoch": 13.022388059701493,
2449
+ "grad_norm": 0.2616482973098755,
2450
+ "learning_rate": 9.582181328841611e-05,
2451
+ "loss": 0.0442,
2452
+ "step": 3490
2453
+ },
2454
+ {
2455
+ "epoch": 13.059701492537313,
2456
+ "grad_norm": 0.20341171324253082,
2457
+ "learning_rate": 9.578866633275288e-05,
2458
+ "loss": 0.0533,
2459
+ "step": 3500
2460
+ },
2461
+ {
2462
+ "epoch": 13.097014925373134,
2463
+ "grad_norm": 0.2223699688911438,
2464
+ "learning_rate": 9.575539419262434e-05,
2465
+ "loss": 0.0458,
2466
+ "step": 3510
2467
+ },
2468
+ {
2469
+ "epoch": 13.134328358208956,
2470
+ "grad_norm": 0.22557464241981506,
2471
+ "learning_rate": 9.572199695899522e-05,
2472
+ "loss": 0.0445,
2473
+ "step": 3520
2474
+ },
2475
+ {
2476
+ "epoch": 13.171641791044776,
2477
+ "grad_norm": 0.25104308128356934,
2478
+ "learning_rate": 9.568847472317232e-05,
2479
+ "loss": 0.0435,
2480
+ "step": 3530
2481
+ },
2482
+ {
2483
+ "epoch": 13.208955223880597,
2484
+ "grad_norm": 0.18720711767673492,
2485
+ "learning_rate": 9.565482757680415e-05,
2486
+ "loss": 0.0453,
2487
+ "step": 3540
2488
+ },
2489
+ {
2490
+ "epoch": 13.246268656716419,
2491
+ "grad_norm": 0.16838951408863068,
2492
+ "learning_rate": 9.562105561188069e-05,
2493
+ "loss": 0.0505,
2494
+ "step": 3550
2495
+ },
2496
+ {
2497
+ "epoch": 13.283582089552239,
2498
+ "grad_norm": 0.31681734323501587,
2499
+ "learning_rate": 9.558715892073323e-05,
2500
+ "loss": 0.0494,
2501
+ "step": 3560
2502
+ },
2503
+ {
2504
+ "epoch": 13.32089552238806,
2505
+ "grad_norm": 0.2390700727701187,
2506
+ "learning_rate": 9.555313759603402e-05,
2507
+ "loss": 0.0538,
2508
+ "step": 3570
2509
+ },
2510
+ {
2511
+ "epoch": 13.35820895522388,
2512
+ "grad_norm": 0.20680709183216095,
2513
+ "learning_rate": 9.551899173079607e-05,
2514
+ "loss": 0.0519,
2515
+ "step": 3580
2516
+ },
2517
+ {
2518
+ "epoch": 13.395522388059701,
2519
+ "grad_norm": 0.2758580148220062,
2520
+ "learning_rate": 9.548472141837286e-05,
2521
+ "loss": 0.0512,
2522
+ "step": 3590
2523
+ },
2524
+ {
2525
+ "epoch": 13.432835820895523,
2526
+ "grad_norm": 0.3653097450733185,
2527
+ "learning_rate": 9.545032675245813e-05,
2528
+ "loss": 0.0496,
2529
+ "step": 3600
2530
+ },
2531
+ {
2532
+ "epoch": 13.470149253731343,
2533
+ "grad_norm": 0.23886866867542267,
2534
+ "learning_rate": 9.541580782708557e-05,
2535
+ "loss": 0.0455,
2536
+ "step": 3610
2537
+ },
2538
+ {
2539
+ "epoch": 13.507462686567164,
2540
+ "grad_norm": 0.3280908465385437,
2541
+ "learning_rate": 9.538116473662861e-05,
2542
+ "loss": 0.0489,
2543
+ "step": 3620
2544
+ },
2545
+ {
2546
+ "epoch": 13.544776119402986,
2547
+ "grad_norm": 0.20268180966377258,
2548
+ "learning_rate": 9.534639757580013e-05,
2549
+ "loss": 0.0484,
2550
+ "step": 3630
2551
+ },
2552
+ {
2553
+ "epoch": 13.582089552238806,
2554
+ "grad_norm": 0.2582015097141266,
2555
+ "learning_rate": 9.531150643965223e-05,
2556
+ "loss": 0.0487,
2557
+ "step": 3640
2558
+ },
2559
+ {
2560
+ "epoch": 13.619402985074627,
2561
+ "grad_norm": 0.18157973885536194,
2562
+ "learning_rate": 9.527649142357596e-05,
2563
+ "loss": 0.0496,
2564
+ "step": 3650
2565
+ },
2566
+ {
2567
+ "epoch": 13.656716417910447,
2568
+ "grad_norm": 0.22841542959213257,
2569
+ "learning_rate": 9.524135262330098e-05,
2570
+ "loss": 0.0467,
2571
+ "step": 3660
2572
+ },
2573
+ {
2574
+ "epoch": 13.694029850746269,
2575
+ "grad_norm": 0.2519935369491577,
2576
+ "learning_rate": 9.520609013489547e-05,
2577
+ "loss": 0.0487,
2578
+ "step": 3670
2579
+ },
2580
+ {
2581
+ "epoch": 13.73134328358209,
2582
+ "grad_norm": 0.24680495262145996,
2583
+ "learning_rate": 9.517070405476575e-05,
2584
+ "loss": 0.0457,
2585
+ "step": 3680
2586
+ },
2587
+ {
2588
+ "epoch": 13.76865671641791,
2589
+ "grad_norm": 0.26362067461013794,
2590
+ "learning_rate": 9.513519447965595e-05,
2591
+ "loss": 0.0495,
2592
+ "step": 3690
2593
+ },
2594
+ {
2595
+ "epoch": 13.805970149253731,
2596
+ "grad_norm": 0.3240712583065033,
2597
+ "learning_rate": 9.509956150664796e-05,
2598
+ "loss": 0.0496,
2599
+ "step": 3700
2600
+ },
2601
+ {
2602
+ "epoch": 13.843283582089553,
2603
+ "grad_norm": 0.21009013056755066,
2604
+ "learning_rate": 9.50638052331609e-05,
2605
+ "loss": 0.0457,
2606
+ "step": 3710
2607
+ },
2608
+ {
2609
+ "epoch": 13.880597014925373,
2610
+ "grad_norm": 0.1669154316186905,
2611
+ "learning_rate": 9.502792575695112e-05,
2612
+ "loss": 0.0496,
2613
+ "step": 3720
2614
+ },
2615
+ {
2616
+ "epoch": 13.917910447761194,
2617
+ "grad_norm": 0.22347605228424072,
2618
+ "learning_rate": 9.499192317611167e-05,
2619
+ "loss": 0.0426,
2620
+ "step": 3730
2621
+ },
2622
+ {
2623
+ "epoch": 13.955223880597014,
2624
+ "grad_norm": 0.15208907425403595,
2625
+ "learning_rate": 9.49557975890723e-05,
2626
+ "loss": 0.0447,
2627
+ "step": 3740
2628
+ },
2629
+ {
2630
+ "epoch": 13.992537313432836,
2631
+ "grad_norm": 0.3206101059913635,
2632
+ "learning_rate": 9.491954909459895e-05,
2633
+ "loss": 0.0471,
2634
+ "step": 3750
2635
+ },
2636
+ {
2637
+ "epoch": 14.029850746268657,
2638
+ "grad_norm": 0.15873713791370392,
2639
+ "learning_rate": 9.488317779179361e-05,
2640
+ "loss": 0.0401,
2641
+ "step": 3760
2642
+ },
2643
+ {
2644
+ "epoch": 14.067164179104477,
2645
+ "grad_norm": 0.19690357148647308,
2646
+ "learning_rate": 9.484668378009408e-05,
2647
+ "loss": 0.0491,
2648
+ "step": 3770
2649
+ },
2650
+ {
2651
+ "epoch": 14.104477611940299,
2652
+ "grad_norm": 0.3211113214492798,
2653
+ "learning_rate": 9.481006715927351e-05,
2654
+ "loss": 0.049,
2655
+ "step": 3780
2656
+ },
2657
+ {
2658
+ "epoch": 14.14179104477612,
2659
+ "grad_norm": 0.27657604217529297,
2660
+ "learning_rate": 9.477332802944044e-05,
2661
+ "loss": 0.0396,
2662
+ "step": 3790
2663
+ },
2664
+ {
2665
+ "epoch": 14.17910447761194,
2666
+ "grad_norm": 0.20194031298160553,
2667
+ "learning_rate": 9.473646649103818e-05,
2668
+ "loss": 0.0442,
2669
+ "step": 3800
2670
+ },
2671
+ {
2672
+ "epoch": 14.216417910447761,
2673
+ "grad_norm": 0.20344595611095428,
2674
+ "learning_rate": 9.46994826448448e-05,
2675
+ "loss": 0.0427,
2676
+ "step": 3810
2677
+ },
2678
+ {
2679
+ "epoch": 14.253731343283581,
2680
+ "grad_norm": 0.2067718505859375,
2681
+ "learning_rate": 9.46623765919727e-05,
2682
+ "loss": 0.0501,
2683
+ "step": 3820
2684
+ },
2685
+ {
2686
+ "epoch": 14.291044776119403,
2687
+ "grad_norm": 0.29719170928001404,
2688
+ "learning_rate": 9.462514843386845e-05,
2689
+ "loss": 0.0519,
2690
+ "step": 3830
2691
+ },
2692
+ {
2693
+ "epoch": 14.328358208955224,
2694
+ "grad_norm": 0.2347182184457779,
2695
+ "learning_rate": 9.458779827231237e-05,
2696
+ "loss": 0.0413,
2697
+ "step": 3840
2698
+ },
2699
+ {
2700
+ "epoch": 14.365671641791044,
2701
+ "grad_norm": 0.1558852344751358,
2702
+ "learning_rate": 9.45503262094184e-05,
2703
+ "loss": 0.0442,
2704
+ "step": 3850
2705
+ },
2706
+ {
2707
+ "epoch": 14.402985074626866,
2708
+ "grad_norm": 0.23085005581378937,
2709
+ "learning_rate": 9.451273234763371e-05,
2710
+ "loss": 0.047,
2711
+ "step": 3860
2712
+ },
2713
+ {
2714
+ "epoch": 14.440298507462687,
2715
+ "grad_norm": 0.1515151560306549,
2716
+ "learning_rate": 9.447501678973852e-05,
2717
+ "loss": 0.0481,
2718
+ "step": 3870
2719
+ },
2720
+ {
2721
+ "epoch": 14.477611940298507,
2722
+ "grad_norm": 0.1916729211807251,
2723
+ "learning_rate": 9.443717963884569e-05,
2724
+ "loss": 0.0474,
2725
+ "step": 3880
2726
+ },
2727
+ {
2728
+ "epoch": 14.514925373134329,
2729
+ "grad_norm": 0.2536492943763733,
2730
+ "learning_rate": 9.439922099840054e-05,
2731
+ "loss": 0.0382,
2732
+ "step": 3890
2733
+ },
2734
+ {
2735
+ "epoch": 14.552238805970148,
2736
+ "grad_norm": 0.1672086864709854,
2737
+ "learning_rate": 9.43611409721806e-05,
2738
+ "loss": 0.0497,
2739
+ "step": 3900
2740
+ },
2741
+ {
2742
+ "epoch": 14.58955223880597,
2743
+ "grad_norm": 0.3644237518310547,
2744
+ "learning_rate": 9.432293966429514e-05,
2745
+ "loss": 0.0444,
2746
+ "step": 3910
2747
+ },
2748
+ {
2749
+ "epoch": 14.626865671641792,
2750
+ "grad_norm": 0.20307251811027527,
2751
+ "learning_rate": 9.428461717918511e-05,
2752
+ "loss": 0.0452,
2753
+ "step": 3920
2754
+ },
2755
+ {
2756
+ "epoch": 14.664179104477611,
2757
+ "grad_norm": 0.20441733300685883,
2758
+ "learning_rate": 9.424617362162271e-05,
2759
+ "loss": 0.0454,
2760
+ "step": 3930
2761
+ },
2762
+ {
2763
+ "epoch": 14.701492537313433,
2764
+ "grad_norm": 0.26315611600875854,
2765
+ "learning_rate": 9.420760909671118e-05,
2766
+ "loss": 0.0486,
2767
+ "step": 3940
2768
+ },
2769
+ {
2770
+ "epoch": 14.738805970149254,
2771
+ "grad_norm": 0.1983092874288559,
2772
+ "learning_rate": 9.416892370988444e-05,
2773
+ "loss": 0.0483,
2774
+ "step": 3950
2775
+ },
2776
+ {
2777
+ "epoch": 14.776119402985074,
2778
+ "grad_norm": 0.18301443755626678,
2779
+ "learning_rate": 9.413011756690685e-05,
2780
+ "loss": 0.0456,
2781
+ "step": 3960
2782
+ },
2783
+ {
2784
+ "epoch": 14.813432835820896,
2785
+ "grad_norm": 0.2433597594499588,
2786
+ "learning_rate": 9.409119077387294e-05,
2787
+ "loss": 0.0463,
2788
+ "step": 3970
2789
+ },
2790
+ {
2791
+ "epoch": 14.850746268656717,
2792
+ "grad_norm": 0.27949392795562744,
2793
+ "learning_rate": 9.405214343720707e-05,
2794
+ "loss": 0.0412,
2795
+ "step": 3980
2796
+ },
2797
+ {
2798
+ "epoch": 14.888059701492537,
2799
+ "grad_norm": 0.22806599736213684,
2800
+ "learning_rate": 9.401297566366318e-05,
2801
+ "loss": 0.0448,
2802
+ "step": 3990
2803
+ },
2804
+ {
2805
+ "epoch": 14.925373134328359,
2806
+ "grad_norm": 0.25421562790870667,
2807
+ "learning_rate": 9.397368756032445e-05,
2808
+ "loss": 0.0426,
2809
+ "step": 4000
2810
+ },
2811
+ {
2812
+ "epoch": 14.962686567164178,
2813
+ "grad_norm": 0.2436474859714508,
2814
+ "learning_rate": 9.393427923460308e-05,
2815
+ "loss": 0.0474,
2816
+ "step": 4010
2817
+ },
2818
+ {
2819
+ "epoch": 15.0,
2820
+ "grad_norm": 0.3756405711174011,
2821
+ "learning_rate": 9.389475079423988e-05,
2822
+ "loss": 0.0438,
2823
+ "step": 4020
2824
+ },
2825
+ {
2826
+ "epoch": 15.037313432835822,
2827
+ "grad_norm": 0.25687697529792786,
2828
+ "learning_rate": 9.385510234730415e-05,
2829
+ "loss": 0.0435,
2830
+ "step": 4030
2831
+ },
2832
+ {
2833
+ "epoch": 15.074626865671641,
2834
+ "grad_norm": 0.17263716459274292,
2835
+ "learning_rate": 9.381533400219318e-05,
2836
+ "loss": 0.0455,
2837
+ "step": 4040
2838
+ },
2839
+ {
2840
+ "epoch": 15.111940298507463,
2841
+ "grad_norm": 0.2471216470003128,
2842
+ "learning_rate": 9.377544586763215e-05,
2843
+ "loss": 0.0429,
2844
+ "step": 4050
2845
+ },
2846
+ {
2847
+ "epoch": 15.149253731343283,
2848
+ "grad_norm": 0.20195460319519043,
2849
+ "learning_rate": 9.373543805267368e-05,
2850
+ "loss": 0.0432,
2851
+ "step": 4060
2852
+ },
2853
+ {
2854
+ "epoch": 15.186567164179104,
2855
+ "grad_norm": 0.1709851622581482,
2856
+ "learning_rate": 9.369531066669758e-05,
2857
+ "loss": 0.0477,
2858
+ "step": 4070
2859
+ },
2860
+ {
2861
+ "epoch": 15.223880597014926,
2862
+ "grad_norm": 0.23063932359218597,
2863
+ "learning_rate": 9.365506381941066e-05,
2864
+ "loss": 0.0379,
2865
+ "step": 4080
2866
+ },
2867
+ {
2868
+ "epoch": 15.261194029850746,
2869
+ "grad_norm": 0.3265426754951477,
2870
+ "learning_rate": 9.36146976208462e-05,
2871
+ "loss": 0.0435,
2872
+ "step": 4090
2873
+ },
2874
+ {
2875
+ "epoch": 15.298507462686567,
2876
+ "grad_norm": 0.26373934745788574,
2877
+ "learning_rate": 9.357421218136386e-05,
2878
+ "loss": 0.047,
2879
+ "step": 4100
2880
+ },
2881
+ {
2882
+ "epoch": 15.335820895522389,
2883
+ "grad_norm": 0.16861388087272644,
2884
+ "learning_rate": 9.353360761164931e-05,
2885
+ "loss": 0.0448,
2886
+ "step": 4110
2887
+ },
2888
+ {
2889
+ "epoch": 15.373134328358208,
2890
+ "grad_norm": 0.303790807723999,
2891
+ "learning_rate": 9.349288402271388e-05,
2892
+ "loss": 0.0396,
2893
+ "step": 4120
2894
+ },
2895
+ {
2896
+ "epoch": 15.41044776119403,
2897
+ "grad_norm": 0.1940719038248062,
2898
+ "learning_rate": 9.345204152589428e-05,
2899
+ "loss": 0.0474,
2900
+ "step": 4130
2901
+ },
2902
+ {
2903
+ "epoch": 15.447761194029852,
2904
+ "grad_norm": 0.34091615676879883,
2905
+ "learning_rate": 9.341108023285238e-05,
2906
+ "loss": 0.0424,
2907
+ "step": 4140
2908
+ },
2909
+ {
2910
+ "epoch": 15.485074626865671,
2911
+ "grad_norm": 0.27036693692207336,
2912
+ "learning_rate": 9.337000025557476e-05,
2913
+ "loss": 0.0482,
2914
+ "step": 4150
2915
+ },
2916
+ {
2917
+ "epoch": 15.522388059701493,
2918
+ "grad_norm": 0.16908007860183716,
2919
+ "learning_rate": 9.332880170637252e-05,
2920
+ "loss": 0.0381,
2921
+ "step": 4160
2922
+ },
2923
+ {
2924
+ "epoch": 15.559701492537313,
2925
+ "grad_norm": 0.23332923650741577,
2926
+ "learning_rate": 9.328748469788093e-05,
2927
+ "loss": 0.0427,
2928
+ "step": 4170
2929
+ },
2930
+ {
2931
+ "epoch": 15.597014925373134,
2932
+ "grad_norm": 0.16899706423282623,
2933
+ "learning_rate": 9.32460493430591e-05,
2934
+ "loss": 0.0439,
2935
+ "step": 4180
2936
+ },
2937
+ {
2938
+ "epoch": 15.634328358208956,
2939
+ "grad_norm": 0.12869524955749512,
2940
+ "learning_rate": 9.320449575518972e-05,
2941
+ "loss": 0.0481,
2942
+ "step": 4190
2943
+ },
2944
+ {
2945
+ "epoch": 15.671641791044776,
2946
+ "grad_norm": 0.21159130334854126,
2947
+ "learning_rate": 9.316282404787871e-05,
2948
+ "loss": 0.0446,
2949
+ "step": 4200
2950
+ },
2951
+ {
2952
+ "epoch": 15.708955223880597,
2953
+ "grad_norm": 0.1849961131811142,
2954
+ "learning_rate": 9.31210343350549e-05,
2955
+ "loss": 0.041,
2956
+ "step": 4210
2957
+ },
2958
+ {
2959
+ "epoch": 15.746268656716419,
2960
+ "grad_norm": 0.16107840836048126,
2961
+ "learning_rate": 9.30791267309698e-05,
2962
+ "loss": 0.0429,
2963
+ "step": 4220
2964
+ },
2965
+ {
2966
+ "epoch": 15.783582089552239,
2967
+ "grad_norm": 0.14206446707248688,
2968
+ "learning_rate": 9.30371013501972e-05,
2969
+ "loss": 0.0409,
2970
+ "step": 4230
2971
+ },
2972
+ {
2973
+ "epoch": 15.82089552238806,
2974
+ "grad_norm": 0.2168441116809845,
2975
+ "learning_rate": 9.299495830763286e-05,
2976
+ "loss": 0.0413,
2977
+ "step": 4240
2978
+ },
2979
+ {
2980
+ "epoch": 15.85820895522388,
2981
+ "grad_norm": 0.21431951224803925,
2982
+ "learning_rate": 9.295269771849427e-05,
2983
+ "loss": 0.0472,
2984
+ "step": 4250
2985
+ },
2986
+ {
2987
+ "epoch": 15.895522388059701,
2988
+ "grad_norm": 0.16851255297660828,
2989
+ "learning_rate": 9.291031969832026e-05,
2990
+ "loss": 0.0508,
2991
+ "step": 4260
2992
+ },
2993
+ {
2994
+ "epoch": 15.932835820895523,
2995
+ "grad_norm": 0.18404732644557953,
2996
+ "learning_rate": 9.286782436297073e-05,
2997
+ "loss": 0.0402,
2998
+ "step": 4270
2999
+ },
3000
+ {
3001
+ "epoch": 15.970149253731343,
3002
+ "grad_norm": 0.21722930669784546,
3003
+ "learning_rate": 9.282521182862629e-05,
3004
+ "loss": 0.0397,
3005
+ "step": 4280
3006
+ },
3007
+ {
3008
+ "epoch": 16.007462686567163,
3009
+ "grad_norm": 0.2523709833621979,
3010
+ "learning_rate": 9.278248221178798e-05,
3011
+ "loss": 0.0427,
3012
+ "step": 4290
3013
+ },
3014
+ {
3015
+ "epoch": 16.044776119402986,
3016
+ "grad_norm": 0.17736563086509705,
3017
+ "learning_rate": 9.273963562927695e-05,
3018
+ "loss": 0.0458,
3019
+ "step": 4300
3020
+ },
3021
+ {
3022
+ "epoch": 16.082089552238806,
3023
+ "grad_norm": 0.20613858103752136,
3024
+ "learning_rate": 9.269667219823412e-05,
3025
+ "loss": 0.0387,
3026
+ "step": 4310
3027
+ },
3028
+ {
3029
+ "epoch": 16.119402985074625,
3030
+ "grad_norm": 0.16557513177394867,
3031
+ "learning_rate": 9.265359203611987e-05,
3032
+ "loss": 0.0411,
3033
+ "step": 4320
3034
+ },
3035
+ {
3036
+ "epoch": 16.15671641791045,
3037
+ "grad_norm": 0.28119519352912903,
3038
+ "learning_rate": 9.261039526071374e-05,
3039
+ "loss": 0.0468,
3040
+ "step": 4330
3041
+ },
3042
+ {
3043
+ "epoch": 16.19402985074627,
3044
+ "grad_norm": 0.21538576483726501,
3045
+ "learning_rate": 9.256708199011401e-05,
3046
+ "loss": 0.0368,
3047
+ "step": 4340
3048
+ },
3049
+ {
3050
+ "epoch": 16.23134328358209,
3051
+ "grad_norm": 0.19657357037067413,
3052
+ "learning_rate": 9.252365234273755e-05,
3053
+ "loss": 0.038,
3054
+ "step": 4350
3055
+ },
3056
+ {
3057
+ "epoch": 16.26865671641791,
3058
+ "grad_norm": 0.19258421659469604,
3059
+ "learning_rate": 9.248010643731935e-05,
3060
+ "loss": 0.0414,
3061
+ "step": 4360
3062
+ },
3063
+ {
3064
+ "epoch": 16.30597014925373,
3065
+ "grad_norm": 0.28801625967025757,
3066
+ "learning_rate": 9.243644439291223e-05,
3067
+ "loss": 0.0387,
3068
+ "step": 4370
3069
+ },
3070
+ {
3071
+ "epoch": 16.34328358208955,
3072
+ "grad_norm": 0.16581468284130096,
3073
+ "learning_rate": 9.239266632888659e-05,
3074
+ "loss": 0.0383,
3075
+ "step": 4380
3076
+ },
3077
+ {
3078
+ "epoch": 16.380597014925375,
3079
+ "grad_norm": 0.34664949774742126,
3080
+ "learning_rate": 9.234877236492997e-05,
3081
+ "loss": 0.0453,
3082
+ "step": 4390
3083
+ },
3084
+ {
3085
+ "epoch": 16.417910447761194,
3086
+ "grad_norm": 0.1439947783946991,
3087
+ "learning_rate": 9.230476262104677e-05,
3088
+ "loss": 0.0466,
3089
+ "step": 4400
3090
+ },
3091
+ {
3092
+ "epoch": 16.455223880597014,
3093
+ "grad_norm": 0.15509940683841705,
3094
+ "learning_rate": 9.226063721755799e-05,
3095
+ "loss": 0.0488,
3096
+ "step": 4410
3097
+ },
3098
+ {
3099
+ "epoch": 16.492537313432837,
3100
+ "grad_norm": 0.18005985021591187,
3101
+ "learning_rate": 9.221639627510076e-05,
3102
+ "loss": 0.0407,
3103
+ "step": 4420
3104
+ },
3105
+ {
3106
+ "epoch": 16.529850746268657,
3107
+ "grad_norm": 0.16012470424175262,
3108
+ "learning_rate": 9.217203991462815e-05,
3109
+ "loss": 0.0394,
3110
+ "step": 4430
3111
+ },
3112
+ {
3113
+ "epoch": 16.567164179104477,
3114
+ "grad_norm": 0.2978847920894623,
3115
+ "learning_rate": 9.212756825740873e-05,
3116
+ "loss": 0.0451,
3117
+ "step": 4440
3118
+ },
3119
+ {
3120
+ "epoch": 16.604477611940297,
3121
+ "grad_norm": 0.2236834019422531,
3122
+ "learning_rate": 9.208298142502636e-05,
3123
+ "loss": 0.0487,
3124
+ "step": 4450
3125
+ },
3126
+ {
3127
+ "epoch": 16.64179104477612,
3128
+ "grad_norm": 0.2686060667037964,
3129
+ "learning_rate": 9.20382795393797e-05,
3130
+ "loss": 0.0403,
3131
+ "step": 4460
3132
+ },
3133
+ {
3134
+ "epoch": 16.67910447761194,
3135
+ "grad_norm": 0.33534038066864014,
3136
+ "learning_rate": 9.199346272268199e-05,
3137
+ "loss": 0.0385,
3138
+ "step": 4470
3139
+ },
3140
+ {
3141
+ "epoch": 16.71641791044776,
3142
+ "grad_norm": 0.19250528514385223,
3143
+ "learning_rate": 9.194853109746074e-05,
3144
+ "loss": 0.0441,
3145
+ "step": 4480
3146
+ },
3147
+ {
3148
+ "epoch": 16.753731343283583,
3149
+ "grad_norm": 0.19218407571315765,
3150
+ "learning_rate": 9.190348478655724e-05,
3151
+ "loss": 0.0474,
3152
+ "step": 4490
3153
+ },
3154
+ {
3155
+ "epoch": 16.791044776119403,
3156
+ "grad_norm": 0.21163488924503326,
3157
+ "learning_rate": 9.185832391312644e-05,
3158
+ "loss": 0.0411,
3159
+ "step": 4500
3160
+ }
3161
+ ],
3162
+ "logging_steps": 10,
3163
+ "max_steps": 20000,
3164
+ "num_input_tokens_seen": 0,
3165
+ "num_train_epochs": 75,
3166
+ "save_steps": 500,
3167
+ "stateful_callbacks": {
3168
+ "TrainerControl": {
3169
+ "args": {
3170
+ "should_epoch_stop": false,
3171
+ "should_evaluate": false,
3172
+ "should_log": false,
3173
+ "should_save": true,
3174
+ "should_training_stop": false
3175
+ },
3176
+ "attributes": {}
3177
+ }
3178
+ },
3179
+ "total_flos": 3.3828481543643136e+17,
3180
+ "train_batch_size": 32,
3181
+ "trial_name": null,
3182
+ "trial_params": null
3183
+ }
checkpoint-5000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-5000/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-5000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:193ed2eff2f1efc1b45d0876f750a171ac697da5510b5d45510990c327aa971e
3
+ size 123328576
checkpoint-5000/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afe44114548aaec257567950fa1c0a2981b6c4d75cd69e6f7a58ecb03c5ab446
3
+ size 246824634
checkpoint-5000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e423bc827d5231f98e5bb1fe4ad825145c336e59988047618a72a0e4330b669e
3
+ size 14244
checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:686df702ebab6ee6c983b8676d5ee13ddeab5e30f320c5136697dbf37d11f69e
3
+ size 1064
checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,3533 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 18.65671641791045,
5
+ "eval_steps": 500,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03731343283582089,
13
+ "grad_norm": 0.8186072111129761,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 1.3847,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.07462686567164178,
20
+ "grad_norm": 0.5007426142692566,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 1.4283,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.11194029850746269,
27
+ "grad_norm": 0.49460887908935547,
28
+ "learning_rate": 3e-06,
29
+ "loss": 1.4868,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.14925373134328357,
34
+ "grad_norm": 0.5032920837402344,
35
+ "learning_rate": 4.000000000000001e-06,
36
+ "loss": 1.4491,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1865671641791045,
41
+ "grad_norm": 0.5688469409942627,
42
+ "learning_rate": 5e-06,
43
+ "loss": 1.3703,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.22388059701492538,
48
+ "grad_norm": 0.5052517652511597,
49
+ "learning_rate": 6e-06,
50
+ "loss": 1.419,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.26119402985074625,
55
+ "grad_norm": 0.6315643787384033,
56
+ "learning_rate": 7.000000000000001e-06,
57
+ "loss": 1.3058,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.29850746268656714,
62
+ "grad_norm": 0.6060447692871094,
63
+ "learning_rate": 8.000000000000001e-06,
64
+ "loss": 1.2908,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3358208955223881,
69
+ "grad_norm": 0.5513179302215576,
70
+ "learning_rate": 9e-06,
71
+ "loss": 1.2311,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.373134328358209,
76
+ "grad_norm": 0.8467404246330261,
77
+ "learning_rate": 1e-05,
78
+ "loss": 1.2043,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.41044776119402987,
83
+ "grad_norm": 0.8141824007034302,
84
+ "learning_rate": 1.1000000000000001e-05,
85
+ "loss": 1.0707,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.44776119402985076,
90
+ "grad_norm": 0.7932347059249878,
91
+ "learning_rate": 1.2e-05,
92
+ "loss": 0.9377,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.48507462686567165,
97
+ "grad_norm": 0.684220552444458,
98
+ "learning_rate": 1.3000000000000001e-05,
99
+ "loss": 0.714,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5223880597014925,
104
+ "grad_norm": 0.5886895060539246,
105
+ "learning_rate": 1.4000000000000001e-05,
106
+ "loss": 0.6479,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5597014925373134,
111
+ "grad_norm": 0.4764939248561859,
112
+ "learning_rate": 1.5e-05,
113
+ "loss": 0.5463,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.5970149253731343,
118
+ "grad_norm": 0.4621008038520813,
119
+ "learning_rate": 1.6000000000000003e-05,
120
+ "loss": 0.4641,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6343283582089553,
125
+ "grad_norm": 0.46492910385131836,
126
+ "learning_rate": 1.7000000000000003e-05,
127
+ "loss": 0.4159,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6716417910447762,
132
+ "grad_norm": 0.5017415881156921,
133
+ "learning_rate": 1.8e-05,
134
+ "loss": 0.4094,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7089552238805971,
139
+ "grad_norm": 0.34392210841178894,
140
+ "learning_rate": 1.9e-05,
141
+ "loss": 0.3478,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.746268656716418,
146
+ "grad_norm": 0.3240516483783722,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3821,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7835820895522388,
153
+ "grad_norm": 0.26301339268684387,
154
+ "learning_rate": 2.1e-05,
155
+ "loss": 0.3606,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8208955223880597,
160
+ "grad_norm": 0.34712520241737366,
161
+ "learning_rate": 2.2000000000000003e-05,
162
+ "loss": 0.3421,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8582089552238806,
167
+ "grad_norm": 0.3248469829559326,
168
+ "learning_rate": 2.3000000000000003e-05,
169
+ "loss": 0.3389,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.8955223880597015,
174
+ "grad_norm": 0.298149436712265,
175
+ "learning_rate": 2.4e-05,
176
+ "loss": 0.3145,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.9328358208955224,
181
+ "grad_norm": 0.2757190763950348,
182
+ "learning_rate": 2.5e-05,
183
+ "loss": 0.3065,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9701492537313433,
188
+ "grad_norm": 0.30510950088500977,
189
+ "learning_rate": 2.6000000000000002e-05,
190
+ "loss": 0.2971,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.007462686567164,
195
+ "grad_norm": 0.37349891662597656,
196
+ "learning_rate": 2.7000000000000002e-05,
197
+ "loss": 0.3273,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.044776119402985,
202
+ "grad_norm": 0.3667634129524231,
203
+ "learning_rate": 2.8000000000000003e-05,
204
+ "loss": 0.308,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.0820895522388059,
209
+ "grad_norm": 0.3463355004787445,
210
+ "learning_rate": 2.9e-05,
211
+ "loss": 0.3109,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.1194029850746268,
216
+ "grad_norm": 0.3888525366783142,
217
+ "learning_rate": 3e-05,
218
+ "loss": 0.2644,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.1567164179104479,
223
+ "grad_norm": 0.3749147951602936,
224
+ "learning_rate": 3.1e-05,
225
+ "loss": 0.2858,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.1940298507462686,
230
+ "grad_norm": 0.3270276188850403,
231
+ "learning_rate": 3.2000000000000005e-05,
232
+ "loss": 0.2573,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.2313432835820897,
237
+ "grad_norm": 0.3658592998981476,
238
+ "learning_rate": 3.3e-05,
239
+ "loss": 0.2613,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.2686567164179103,
244
+ "grad_norm": 0.3526328206062317,
245
+ "learning_rate": 3.4000000000000007e-05,
246
+ "loss": 0.2328,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.3059701492537314,
251
+ "grad_norm": 0.4528139531612396,
252
+ "learning_rate": 3.5e-05,
253
+ "loss": 0.2429,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.3432835820895521,
258
+ "grad_norm": 0.5426791310310364,
259
+ "learning_rate": 3.6e-05,
260
+ "loss": 0.2209,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.3805970149253732,
265
+ "grad_norm": 0.41844552755355835,
266
+ "learning_rate": 3.7e-05,
267
+ "loss": 0.2319,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.417910447761194,
272
+ "grad_norm": 0.4749431908130646,
273
+ "learning_rate": 3.8e-05,
274
+ "loss": 0.2233,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.455223880597015,
279
+ "grad_norm": 0.7010189890861511,
280
+ "learning_rate": 3.9000000000000006e-05,
281
+ "loss": 0.2181,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.4925373134328357,
286
+ "grad_norm": 0.5747635960578918,
287
+ "learning_rate": 4e-05,
288
+ "loss": 0.213,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.5298507462686568,
293
+ "grad_norm": 0.3661474287509918,
294
+ "learning_rate": 4.1e-05,
295
+ "loss": 0.2171,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.5671641791044775,
300
+ "grad_norm": 0.467835396528244,
301
+ "learning_rate": 4.2e-05,
302
+ "loss": 0.1985,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.6044776119402986,
307
+ "grad_norm": 0.5470123291015625,
308
+ "learning_rate": 4.3e-05,
309
+ "loss": 0.2176,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.6417910447761193,
314
+ "grad_norm": 0.5761199593544006,
315
+ "learning_rate": 4.4000000000000006e-05,
316
+ "loss": 0.2007,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.6791044776119404,
321
+ "grad_norm": 0.48257485032081604,
322
+ "learning_rate": 4.5e-05,
323
+ "loss": 0.2043,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.716417910447761,
328
+ "grad_norm": 0.48353052139282227,
329
+ "learning_rate": 4.600000000000001e-05,
330
+ "loss": 0.1872,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.7537313432835822,
335
+ "grad_norm": 0.4388391375541687,
336
+ "learning_rate": 4.7e-05,
337
+ "loss": 0.206,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.7910447761194028,
342
+ "grad_norm": 0.47332626581192017,
343
+ "learning_rate": 4.8e-05,
344
+ "loss": 0.1876,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.828358208955224,
349
+ "grad_norm": 0.8053535223007202,
350
+ "learning_rate": 4.9e-05,
351
+ "loss": 0.1839,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 1.8656716417910446,
356
+ "grad_norm": 0.413979709148407,
357
+ "learning_rate": 5e-05,
358
+ "loss": 0.1732,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 1.9029850746268657,
363
+ "grad_norm": 0.36910712718963623,
364
+ "learning_rate": 5.1000000000000006e-05,
365
+ "loss": 0.1827,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 1.9402985074626866,
370
+ "grad_norm": 0.8458298444747925,
371
+ "learning_rate": 5.2000000000000004e-05,
372
+ "loss": 0.1727,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 1.9776119402985075,
377
+ "grad_norm": 0.5452115535736084,
378
+ "learning_rate": 5.300000000000001e-05,
379
+ "loss": 0.1818,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.014925373134328,
384
+ "grad_norm": 0.4518108069896698,
385
+ "learning_rate": 5.4000000000000005e-05,
386
+ "loss": 0.177,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.0522388059701493,
391
+ "grad_norm": 0.66865074634552,
392
+ "learning_rate": 5.500000000000001e-05,
393
+ "loss": 0.1726,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.08955223880597,
398
+ "grad_norm": 0.6536034345626831,
399
+ "learning_rate": 5.6000000000000006e-05,
400
+ "loss": 0.1541,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.126865671641791,
405
+ "grad_norm": 0.5571377277374268,
406
+ "learning_rate": 5.6999999999999996e-05,
407
+ "loss": 0.1671,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.1641791044776117,
412
+ "grad_norm": 0.5385546684265137,
413
+ "learning_rate": 5.8e-05,
414
+ "loss": 0.1582,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.201492537313433,
419
+ "grad_norm": 0.577961266040802,
420
+ "learning_rate": 5.9e-05,
421
+ "loss": 0.1528,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.2388059701492535,
426
+ "grad_norm": 0.5082416534423828,
427
+ "learning_rate": 6e-05,
428
+ "loss": 0.1638,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.2761194029850746,
433
+ "grad_norm": 0.5490861535072327,
434
+ "learning_rate": 6.1e-05,
435
+ "loss": 0.166,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.3134328358208958,
440
+ "grad_norm": 0.492366760969162,
441
+ "learning_rate": 6.2e-05,
442
+ "loss": 0.1481,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.3507462686567164,
447
+ "grad_norm": 0.3702855110168457,
448
+ "learning_rate": 6.3e-05,
449
+ "loss": 0.1514,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.388059701492537,
454
+ "grad_norm": 0.664667010307312,
455
+ "learning_rate": 6.400000000000001e-05,
456
+ "loss": 0.1441,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.425373134328358,
461
+ "grad_norm": 0.33382174372673035,
462
+ "learning_rate": 6.500000000000001e-05,
463
+ "loss": 0.1573,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.4626865671641793,
468
+ "grad_norm": 0.4848814010620117,
469
+ "learning_rate": 6.6e-05,
470
+ "loss": 0.1457,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.5,
475
+ "grad_norm": 0.3649997413158417,
476
+ "learning_rate": 6.7e-05,
477
+ "loss": 0.1467,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.5373134328358207,
482
+ "grad_norm": 0.6385223865509033,
483
+ "learning_rate": 6.800000000000001e-05,
484
+ "loss": 0.145,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.574626865671642,
489
+ "grad_norm": 0.4580625891685486,
490
+ "learning_rate": 6.9e-05,
491
+ "loss": 0.1352,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.611940298507463,
496
+ "grad_norm": 0.5141746401786804,
497
+ "learning_rate": 7e-05,
498
+ "loss": 0.1444,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.6492537313432836,
503
+ "grad_norm": 0.40220722556114197,
504
+ "learning_rate": 7.1e-05,
505
+ "loss": 0.1493,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.6865671641791042,
510
+ "grad_norm": 0.5510571002960205,
511
+ "learning_rate": 7.2e-05,
512
+ "loss": 0.1387,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.7238805970149254,
517
+ "grad_norm": 0.43814659118652344,
518
+ "learning_rate": 7.3e-05,
519
+ "loss": 0.1374,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 2.7611940298507465,
524
+ "grad_norm": 0.4118008613586426,
525
+ "learning_rate": 7.4e-05,
526
+ "loss": 0.1297,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 2.798507462686567,
531
+ "grad_norm": 0.5626503229141235,
532
+ "learning_rate": 7.500000000000001e-05,
533
+ "loss": 0.1299,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 2.835820895522388,
538
+ "grad_norm": 0.4066360592842102,
539
+ "learning_rate": 7.6e-05,
540
+ "loss": 0.1102,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 2.873134328358209,
545
+ "grad_norm": 0.47184985876083374,
546
+ "learning_rate": 7.7e-05,
547
+ "loss": 0.1219,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 2.91044776119403,
552
+ "grad_norm": 0.6611475348472595,
553
+ "learning_rate": 7.800000000000001e-05,
554
+ "loss": 0.1267,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 2.9477611940298507,
559
+ "grad_norm": 0.3570108413696289,
560
+ "learning_rate": 7.900000000000001e-05,
561
+ "loss": 0.1191,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 2.9850746268656714,
566
+ "grad_norm": 0.4581681489944458,
567
+ "learning_rate": 8e-05,
568
+ "loss": 0.1209,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.0223880597014925,
573
+ "grad_norm": 0.4643435776233673,
574
+ "learning_rate": 8.1e-05,
575
+ "loss": 0.129,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.0597014925373136,
580
+ "grad_norm": 0.5595763921737671,
581
+ "learning_rate": 8.2e-05,
582
+ "loss": 0.1158,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.0970149253731343,
587
+ "grad_norm": 0.48848605155944824,
588
+ "learning_rate": 8.3e-05,
589
+ "loss": 0.1188,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.1343283582089554,
594
+ "grad_norm": 0.4496570825576782,
595
+ "learning_rate": 8.4e-05,
596
+ "loss": 0.114,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.171641791044776,
601
+ "grad_norm": 0.31364986300468445,
602
+ "learning_rate": 8.5e-05,
603
+ "loss": 0.1196,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.208955223880597,
608
+ "grad_norm": 0.3395878076553345,
609
+ "learning_rate": 8.6e-05,
610
+ "loss": 0.1124,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.246268656716418,
615
+ "grad_norm": 0.4917413592338562,
616
+ "learning_rate": 8.7e-05,
617
+ "loss": 0.1074,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.283582089552239,
622
+ "grad_norm": 0.44114553928375244,
623
+ "learning_rate": 8.800000000000001e-05,
624
+ "loss": 0.1095,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.3208955223880596,
629
+ "grad_norm": 0.3323831558227539,
630
+ "learning_rate": 8.900000000000001e-05,
631
+ "loss": 0.106,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.3582089552238807,
636
+ "grad_norm": 0.4495660066604614,
637
+ "learning_rate": 9e-05,
638
+ "loss": 0.1222,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.3955223880597014,
643
+ "grad_norm": 0.40784788131713867,
644
+ "learning_rate": 9.1e-05,
645
+ "loss": 0.1048,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.4328358208955225,
650
+ "grad_norm": 0.4643700420856476,
651
+ "learning_rate": 9.200000000000001e-05,
652
+ "loss": 0.1097,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.470149253731343,
657
+ "grad_norm": 0.472494512796402,
658
+ "learning_rate": 9.300000000000001e-05,
659
+ "loss": 0.1041,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.5074626865671643,
664
+ "grad_norm": 0.6110897660255432,
665
+ "learning_rate": 9.4e-05,
666
+ "loss": 0.0959,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.544776119402985,
671
+ "grad_norm": 0.5313069820404053,
672
+ "learning_rate": 9.5e-05,
673
+ "loss": 0.113,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.582089552238806,
678
+ "grad_norm": 0.4223133623600006,
679
+ "learning_rate": 9.6e-05,
680
+ "loss": 0.099,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.6194029850746268,
685
+ "grad_norm": 0.5464731454849243,
686
+ "learning_rate": 9.7e-05,
687
+ "loss": 0.1008,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 3.656716417910448,
692
+ "grad_norm": 0.3538314402103424,
693
+ "learning_rate": 9.8e-05,
694
+ "loss": 0.1049,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 3.6940298507462686,
699
+ "grad_norm": 0.7460148334503174,
700
+ "learning_rate": 9.900000000000001e-05,
701
+ "loss": 0.1088,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 3.7313432835820897,
706
+ "grad_norm": 0.3210597038269043,
707
+ "learning_rate": 0.0001,
708
+ "loss": 0.1041,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 3.7686567164179103,
713
+ "grad_norm": 0.4450497627258301,
714
+ "learning_rate": 9.999993165095463e-05,
715
+ "loss": 0.0985,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 3.8059701492537314,
720
+ "grad_norm": 0.4348960816860199,
721
+ "learning_rate": 9.999972660400536e-05,
722
+ "loss": 0.1015,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 3.843283582089552,
727
+ "grad_norm": 0.462782621383667,
728
+ "learning_rate": 9.999938485971279e-05,
729
+ "loss": 0.1068,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 3.8805970149253732,
734
+ "grad_norm": 0.3801368474960327,
735
+ "learning_rate": 9.999890641901125e-05,
736
+ "loss": 0.1117,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 3.917910447761194,
741
+ "grad_norm": 0.45135366916656494,
742
+ "learning_rate": 9.999829128320874e-05,
743
+ "loss": 0.0917,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 3.955223880597015,
748
+ "grad_norm": 0.41138389706611633,
749
+ "learning_rate": 9.999753945398704e-05,
750
+ "loss": 0.1049,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 3.9925373134328357,
755
+ "grad_norm": 0.4976252317428589,
756
+ "learning_rate": 9.999665093340165e-05,
757
+ "loss": 0.1029,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.029850746268656,
762
+ "grad_norm": 0.46372008323669434,
763
+ "learning_rate": 9.99956257238817e-05,
764
+ "loss": 0.1012,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.067164179104478,
769
+ "grad_norm": 0.546938955783844,
770
+ "learning_rate": 9.999446382823013e-05,
771
+ "loss": 0.0829,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.104477611940299,
776
+ "grad_norm": 0.40513405203819275,
777
+ "learning_rate": 9.999316524962345e-05,
778
+ "loss": 0.0933,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.141791044776119,
783
+ "grad_norm": 0.4198484420776367,
784
+ "learning_rate": 9.999172999161198e-05,
785
+ "loss": 0.0895,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.17910447761194,
790
+ "grad_norm": 0.3965628743171692,
791
+ "learning_rate": 9.999015805811965e-05,
792
+ "loss": 0.0917,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.2164179104477615,
797
+ "grad_norm": 0.3095884621143341,
798
+ "learning_rate": 9.998844945344405e-05,
799
+ "loss": 0.0953,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.253731343283582,
804
+ "grad_norm": 0.7962276339530945,
805
+ "learning_rate": 9.998660418225645e-05,
806
+ "loss": 0.0979,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.291044776119403,
811
+ "grad_norm": 0.42066490650177,
812
+ "learning_rate": 9.998462224960175e-05,
813
+ "loss": 0.099,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.3283582089552235,
818
+ "grad_norm": 0.3894193470478058,
819
+ "learning_rate": 9.998250366089848e-05,
820
+ "loss": 0.0887,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.365671641791045,
825
+ "grad_norm": 0.28998032212257385,
826
+ "learning_rate": 9.998024842193876e-05,
827
+ "loss": 0.0943,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.402985074626866,
832
+ "grad_norm": 0.3919823467731476,
833
+ "learning_rate": 9.997785653888835e-05,
834
+ "loss": 0.0916,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.440298507462686,
839
+ "grad_norm": 0.3708650469779968,
840
+ "learning_rate": 9.997532801828658e-05,
841
+ "loss": 0.0858,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.477611940298507,
846
+ "grad_norm": 0.2935069799423218,
847
+ "learning_rate": 9.997266286704631e-05,
848
+ "loss": 0.0992,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.514925373134329,
853
+ "grad_norm": 0.4675377607345581,
854
+ "learning_rate": 9.996986109245395e-05,
855
+ "loss": 0.0854,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.552238805970149,
860
+ "grad_norm": 0.31374865770339966,
861
+ "learning_rate": 9.996692270216947e-05,
862
+ "loss": 0.0788,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 4.58955223880597,
867
+ "grad_norm": 0.419249951839447,
868
+ "learning_rate": 9.996384770422629e-05,
869
+ "loss": 0.0873,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 4.6268656716417915,
874
+ "grad_norm": 0.26002731919288635,
875
+ "learning_rate": 9.996063610703137e-05,
876
+ "loss": 0.0845,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 4.664179104477612,
881
+ "grad_norm": 0.29573896527290344,
882
+ "learning_rate": 9.995728791936504e-05,
883
+ "loss": 0.091,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 4.701492537313433,
888
+ "grad_norm": 0.33090147376060486,
889
+ "learning_rate": 9.995380315038119e-05,
890
+ "loss": 0.0827,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 4.7388059701492535,
895
+ "grad_norm": 0.24417485296726227,
896
+ "learning_rate": 9.9950181809607e-05,
897
+ "loss": 0.0859,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 4.776119402985074,
902
+ "grad_norm": 0.48290401697158813,
903
+ "learning_rate": 9.994642390694308e-05,
904
+ "loss": 0.0889,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 4.813432835820896,
909
+ "grad_norm": 0.4479697048664093,
910
+ "learning_rate": 9.99425294526634e-05,
911
+ "loss": 0.097,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 4.850746268656716,
916
+ "grad_norm": 0.3560147285461426,
917
+ "learning_rate": 9.993849845741524e-05,
918
+ "loss": 0.0904,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 4.888059701492537,
923
+ "grad_norm": 0.6645416617393494,
924
+ "learning_rate": 9.99343309322192e-05,
925
+ "loss": 0.0922,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 4.925373134328359,
930
+ "grad_norm": 0.29696759581565857,
931
+ "learning_rate": 9.993002688846913e-05,
932
+ "loss": 0.093,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 4.962686567164179,
937
+ "grad_norm": 0.47146692872047424,
938
+ "learning_rate": 9.992558633793212e-05,
939
+ "loss": 0.085,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.0,
944
+ "grad_norm": 0.3430916368961334,
945
+ "learning_rate": 9.992100929274846e-05,
946
+ "loss": 0.0805,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.037313432835821,
951
+ "grad_norm": 0.3205055892467499,
952
+ "learning_rate": 9.991629576543163e-05,
953
+ "loss": 0.0766,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.074626865671641,
958
+ "grad_norm": 0.3664805293083191,
959
+ "learning_rate": 9.991144576886823e-05,
960
+ "loss": 0.0766,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.111940298507463,
965
+ "grad_norm": 0.3753412663936615,
966
+ "learning_rate": 9.990645931631796e-05,
967
+ "loss": 0.0688,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.149253731343284,
972
+ "grad_norm": 0.31633055210113525,
973
+ "learning_rate": 9.990133642141359e-05,
974
+ "loss": 0.0796,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.186567164179104,
979
+ "grad_norm": 0.3355732262134552,
980
+ "learning_rate": 9.989607709816091e-05,
981
+ "loss": 0.0716,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.223880597014926,
986
+ "grad_norm": 0.24850831925868988,
987
+ "learning_rate": 9.989068136093873e-05,
988
+ "loss": 0.0778,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.2611940298507465,
993
+ "grad_norm": 0.29537102580070496,
994
+ "learning_rate": 9.988514922449879e-05,
995
+ "loss": 0.0759,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.298507462686567,
1000
+ "grad_norm": 0.3430945873260498,
1001
+ "learning_rate": 9.987948070396571e-05,
1002
+ "loss": 0.0774,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.335820895522388,
1007
+ "grad_norm": 0.5220637917518616,
1008
+ "learning_rate": 9.987367581483705e-05,
1009
+ "loss": 0.0836,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.373134328358209,
1014
+ "grad_norm": 0.28184008598327637,
1015
+ "learning_rate": 9.986773457298311e-05,
1016
+ "loss": 0.0752,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.41044776119403,
1021
+ "grad_norm": 0.36261311173439026,
1022
+ "learning_rate": 9.986165699464705e-05,
1023
+ "loss": 0.075,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.447761194029851,
1028
+ "grad_norm": 0.5107380151748657,
1029
+ "learning_rate": 9.985544309644475e-05,
1030
+ "loss": 0.0814,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 5.485074626865671,
1035
+ "grad_norm": 0.2446671426296234,
1036
+ "learning_rate": 9.984909289536473e-05,
1037
+ "loss": 0.0704,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 5.522388059701493,
1042
+ "grad_norm": 0.30449381470680237,
1043
+ "learning_rate": 9.984260640876821e-05,
1044
+ "loss": 0.0794,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 5.559701492537314,
1049
+ "grad_norm": 0.25645050406455994,
1050
+ "learning_rate": 9.983598365438902e-05,
1051
+ "loss": 0.0709,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 5.597014925373134,
1056
+ "grad_norm": 0.23825006186962128,
1057
+ "learning_rate": 9.98292246503335e-05,
1058
+ "loss": 0.0828,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 5.634328358208955,
1063
+ "grad_norm": 0.3259269893169403,
1064
+ "learning_rate": 9.98223294150805e-05,
1065
+ "loss": 0.0824,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 5.6716417910447765,
1070
+ "grad_norm": 0.24058914184570312,
1071
+ "learning_rate": 9.981529796748134e-05,
1072
+ "loss": 0.073,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 5.708955223880597,
1077
+ "grad_norm": 0.34457242488861084,
1078
+ "learning_rate": 9.980813032675974e-05,
1079
+ "loss": 0.0845,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 5.746268656716418,
1084
+ "grad_norm": 0.32940393686294556,
1085
+ "learning_rate": 9.980082651251175e-05,
1086
+ "loss": 0.0832,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 5.7835820895522385,
1091
+ "grad_norm": 0.5683007836341858,
1092
+ "learning_rate": 9.979338654470569e-05,
1093
+ "loss": 0.0836,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 5.82089552238806,
1098
+ "grad_norm": 0.31041061878204346,
1099
+ "learning_rate": 9.97858104436822e-05,
1100
+ "loss": 0.07,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 5.858208955223881,
1105
+ "grad_norm": 0.37858131527900696,
1106
+ "learning_rate": 9.977809823015401e-05,
1107
+ "loss": 0.0738,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 5.895522388059701,
1112
+ "grad_norm": 0.2743091583251953,
1113
+ "learning_rate": 9.977024992520602e-05,
1114
+ "loss": 0.0761,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 5.932835820895522,
1119
+ "grad_norm": 0.29117098450660706,
1120
+ "learning_rate": 9.976226555029522e-05,
1121
+ "loss": 0.0777,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 5.970149253731344,
1126
+ "grad_norm": 0.31398633122444153,
1127
+ "learning_rate": 9.975414512725057e-05,
1128
+ "loss": 0.0664,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.007462686567164,
1133
+ "grad_norm": 0.2684272527694702,
1134
+ "learning_rate": 9.974588867827301e-05,
1135
+ "loss": 0.0686,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.044776119402985,
1140
+ "grad_norm": 0.3945397436618805,
1141
+ "learning_rate": 9.973749622593534e-05,
1142
+ "loss": 0.0614,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.082089552238806,
1147
+ "grad_norm": 0.2747954726219177,
1148
+ "learning_rate": 9.972896779318219e-05,
1149
+ "loss": 0.0681,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.119402985074627,
1154
+ "grad_norm": 0.43257200717926025,
1155
+ "learning_rate": 9.972030340333001e-05,
1156
+ "loss": 0.0725,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.156716417910448,
1161
+ "grad_norm": 0.3559250831604004,
1162
+ "learning_rate": 9.97115030800669e-05,
1163
+ "loss": 0.0804,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.1940298507462686,
1168
+ "grad_norm": 0.3079264760017395,
1169
+ "learning_rate": 9.970256684745258e-05,
1170
+ "loss": 0.0649,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.231343283582089,
1175
+ "grad_norm": 0.32298946380615234,
1176
+ "learning_rate": 9.969349472991838e-05,
1177
+ "loss": 0.0668,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.268656716417911,
1182
+ "grad_norm": 0.2826225459575653,
1183
+ "learning_rate": 9.968428675226714e-05,
1184
+ "loss": 0.0734,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.3059701492537314,
1189
+ "grad_norm": 0.39002349972724915,
1190
+ "learning_rate": 9.967494293967312e-05,
1191
+ "loss": 0.0728,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.343283582089552,
1196
+ "grad_norm": 0.403890997171402,
1197
+ "learning_rate": 9.966546331768191e-05,
1198
+ "loss": 0.067,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.380597014925373,
1203
+ "grad_norm": 0.3755359351634979,
1204
+ "learning_rate": 9.965584791221048e-05,
1205
+ "loss": 0.0755,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 6.417910447761194,
1210
+ "grad_norm": 0.26346635818481445,
1211
+ "learning_rate": 9.964609674954696e-05,
1212
+ "loss": 0.0728,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 6.455223880597015,
1217
+ "grad_norm": 0.45292145013809204,
1218
+ "learning_rate": 9.963620985635065e-05,
1219
+ "loss": 0.0731,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 6.492537313432836,
1224
+ "grad_norm": 0.3568434715270996,
1225
+ "learning_rate": 9.962618725965196e-05,
1226
+ "loss": 0.0761,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 6.529850746268656,
1231
+ "grad_norm": 0.2551257014274597,
1232
+ "learning_rate": 9.961602898685226e-05,
1233
+ "loss": 0.0694,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 6.567164179104478,
1238
+ "grad_norm": 0.6106354594230652,
1239
+ "learning_rate": 9.96057350657239e-05,
1240
+ "loss": 0.0827,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 6.604477611940299,
1245
+ "grad_norm": 0.3226093053817749,
1246
+ "learning_rate": 9.959530552441005e-05,
1247
+ "loss": 0.0716,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 6.641791044776119,
1252
+ "grad_norm": 0.4297254979610443,
1253
+ "learning_rate": 9.95847403914247e-05,
1254
+ "loss": 0.0748,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 6.67910447761194,
1259
+ "grad_norm": 0.26469680666923523,
1260
+ "learning_rate": 9.95740396956525e-05,
1261
+ "loss": 0.074,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 6.7164179104477615,
1266
+ "grad_norm": 0.22717897593975067,
1267
+ "learning_rate": 9.956320346634876e-05,
1268
+ "loss": 0.0739,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 6.753731343283582,
1273
+ "grad_norm": 0.4513498544692993,
1274
+ "learning_rate": 9.955223173313931e-05,
1275
+ "loss": 0.0664,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 6.791044776119403,
1280
+ "grad_norm": 0.31683439016342163,
1281
+ "learning_rate": 9.954112452602045e-05,
1282
+ "loss": 0.069,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 6.8283582089552235,
1287
+ "grad_norm": 0.3350532650947571,
1288
+ "learning_rate": 9.952988187535886e-05,
1289
+ "loss": 0.0699,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 6.865671641791045,
1294
+ "grad_norm": 0.29829463362693787,
1295
+ "learning_rate": 9.95185038118915e-05,
1296
+ "loss": 0.0663,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 6.902985074626866,
1301
+ "grad_norm": 0.31650781631469727,
1302
+ "learning_rate": 9.950699036672559e-05,
1303
+ "loss": 0.0668,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 6.940298507462686,
1308
+ "grad_norm": 0.360944926738739,
1309
+ "learning_rate": 9.949534157133844e-05,
1310
+ "loss": 0.0696,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 6.977611940298507,
1315
+ "grad_norm": 0.31337013840675354,
1316
+ "learning_rate": 9.948355745757741e-05,
1317
+ "loss": 0.073,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.014925373134329,
1322
+ "grad_norm": 0.4675919711589813,
1323
+ "learning_rate": 9.94716380576598e-05,
1324
+ "loss": 0.0688,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.052238805970149,
1329
+ "grad_norm": 0.3031919002532959,
1330
+ "learning_rate": 9.945958340417283e-05,
1331
+ "loss": 0.0596,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.08955223880597,
1336
+ "grad_norm": 0.24858474731445312,
1337
+ "learning_rate": 9.944739353007344e-05,
1338
+ "loss": 0.0717,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.126865671641791,
1343
+ "grad_norm": 0.20959483087062836,
1344
+ "learning_rate": 9.943506846868826e-05,
1345
+ "loss": 0.0694,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.164179104477612,
1350
+ "grad_norm": 0.35621434450149536,
1351
+ "learning_rate": 9.942260825371358e-05,
1352
+ "loss": 0.063,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.201492537313433,
1357
+ "grad_norm": 0.3462587594985962,
1358
+ "learning_rate": 9.941001291921512e-05,
1359
+ "loss": 0.068,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.2388059701492535,
1364
+ "grad_norm": 0.38649681210517883,
1365
+ "learning_rate": 9.939728249962807e-05,
1366
+ "loss": 0.0638,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.276119402985074,
1371
+ "grad_norm": 0.29564595222473145,
1372
+ "learning_rate": 9.938441702975689e-05,
1373
+ "loss": 0.0626,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 7.313432835820896,
1378
+ "grad_norm": 0.339857816696167,
1379
+ "learning_rate": 9.937141654477528e-05,
1380
+ "loss": 0.0535,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 7.350746268656716,
1385
+ "grad_norm": 0.2591215670108795,
1386
+ "learning_rate": 9.93582810802261e-05,
1387
+ "loss": 0.0645,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 7.388059701492537,
1392
+ "grad_norm": 0.30237796902656555,
1393
+ "learning_rate": 9.934501067202117e-05,
1394
+ "loss": 0.0675,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 7.425373134328359,
1399
+ "grad_norm": 0.28394174575805664,
1400
+ "learning_rate": 9.93316053564413e-05,
1401
+ "loss": 0.0643,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 7.462686567164179,
1406
+ "grad_norm": 0.3124663233757019,
1407
+ "learning_rate": 9.931806517013612e-05,
1408
+ "loss": 0.059,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 7.5,
1413
+ "grad_norm": 0.36073037981987,
1414
+ "learning_rate": 9.930439015012396e-05,
1415
+ "loss": 0.0606,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 7.537313432835821,
1420
+ "grad_norm": 0.4091481864452362,
1421
+ "learning_rate": 9.929058033379181e-05,
1422
+ "loss": 0.0603,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 7.574626865671641,
1427
+ "grad_norm": 0.44718074798583984,
1428
+ "learning_rate": 9.927663575889521e-05,
1429
+ "loss": 0.0741,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 7.611940298507463,
1434
+ "grad_norm": 0.3819601833820343,
1435
+ "learning_rate": 9.926255646355804e-05,
1436
+ "loss": 0.0707,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 7.649253731343284,
1441
+ "grad_norm": 0.23336420953273773,
1442
+ "learning_rate": 9.92483424862726e-05,
1443
+ "loss": 0.0676,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 7.686567164179104,
1448
+ "grad_norm": 0.24415315687656403,
1449
+ "learning_rate": 9.923399386589933e-05,
1450
+ "loss": 0.0594,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 7.723880597014926,
1455
+ "grad_norm": 0.3735473155975342,
1456
+ "learning_rate": 9.921951064166684e-05,
1457
+ "loss": 0.062,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 7.7611940298507465,
1462
+ "grad_norm": 0.31629472970962524,
1463
+ "learning_rate": 9.92048928531717e-05,
1464
+ "loss": 0.0606,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 7.798507462686567,
1469
+ "grad_norm": 0.37902557849884033,
1470
+ "learning_rate": 9.919014054037836e-05,
1471
+ "loss": 0.0584,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 7.835820895522388,
1476
+ "grad_norm": 0.3486720323562622,
1477
+ "learning_rate": 9.917525374361912e-05,
1478
+ "loss": 0.056,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 7.8731343283582085,
1483
+ "grad_norm": 0.3731362521648407,
1484
+ "learning_rate": 9.91602325035939e-05,
1485
+ "loss": 0.0601,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 7.91044776119403,
1490
+ "grad_norm": 0.3560399115085602,
1491
+ "learning_rate": 9.914507686137019e-05,
1492
+ "loss": 0.06,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 7.947761194029851,
1497
+ "grad_norm": 0.30075564980506897,
1498
+ "learning_rate": 9.912978685838294e-05,
1499
+ "loss": 0.0657,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 7.985074626865671,
1504
+ "grad_norm": 0.2984028458595276,
1505
+ "learning_rate": 9.911436253643445e-05,
1506
+ "loss": 0.0587,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 8.022388059701493,
1511
+ "grad_norm": 0.1980169117450714,
1512
+ "learning_rate": 9.90988039376942e-05,
1513
+ "loss": 0.0718,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 8.059701492537313,
1518
+ "grad_norm": 0.31339579820632935,
1519
+ "learning_rate": 9.90831111046988e-05,
1520
+ "loss": 0.0557,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 8.097014925373134,
1525
+ "grad_norm": 0.1968696266412735,
1526
+ "learning_rate": 9.90672840803519e-05,
1527
+ "loss": 0.0571,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 8.134328358208956,
1532
+ "grad_norm": 0.23931682109832764,
1533
+ "learning_rate": 9.905132290792394e-05,
1534
+ "loss": 0.0566,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 8.171641791044776,
1539
+ "grad_norm": 0.21741189062595367,
1540
+ "learning_rate": 9.903522763105218e-05,
1541
+ "loss": 0.0575,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 8.208955223880597,
1546
+ "grad_norm": 0.22874368727207184,
1547
+ "learning_rate": 9.901899829374047e-05,
1548
+ "loss": 0.0565,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 8.246268656716419,
1553
+ "grad_norm": 0.3441888093948364,
1554
+ "learning_rate": 9.900263494035921e-05,
1555
+ "loss": 0.0565,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 8.283582089552239,
1560
+ "grad_norm": 0.2539830803871155,
1561
+ "learning_rate": 9.89861376156452e-05,
1562
+ "loss": 0.0538,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 8.32089552238806,
1567
+ "grad_norm": 0.2235102653503418,
1568
+ "learning_rate": 9.896950636470147e-05,
1569
+ "loss": 0.0609,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 8.35820895522388,
1574
+ "grad_norm": 0.1941322684288025,
1575
+ "learning_rate": 9.895274123299723e-05,
1576
+ "loss": 0.0562,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 8.395522388059701,
1581
+ "grad_norm": 0.2691369950771332,
1582
+ "learning_rate": 9.893584226636772e-05,
1583
+ "loss": 0.0608,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 8.432835820895523,
1588
+ "grad_norm": 0.24730461835861206,
1589
+ "learning_rate": 9.891880951101407e-05,
1590
+ "loss": 0.0582,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 8.470149253731343,
1595
+ "grad_norm": 0.34785839915275574,
1596
+ "learning_rate": 9.890164301350318e-05,
1597
+ "loss": 0.0506,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 8.507462686567164,
1602
+ "grad_norm": 0.3625825345516205,
1603
+ "learning_rate": 9.888434282076758e-05,
1604
+ "loss": 0.0614,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 8.544776119402986,
1609
+ "grad_norm": 0.25210148096084595,
1610
+ "learning_rate": 9.886690898010535e-05,
1611
+ "loss": 0.0611,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 8.582089552238806,
1616
+ "grad_norm": 0.27312466502189636,
1617
+ "learning_rate": 9.884934153917997e-05,
1618
+ "loss": 0.0537,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 8.619402985074627,
1623
+ "grad_norm": 0.314647912979126,
1624
+ "learning_rate": 9.883164054602012e-05,
1625
+ "loss": 0.0602,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 8.656716417910447,
1630
+ "grad_norm": 0.21531912684440613,
1631
+ "learning_rate": 9.881380604901964e-05,
1632
+ "loss": 0.0552,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 8.694029850746269,
1637
+ "grad_norm": 0.23920664191246033,
1638
+ "learning_rate": 9.879583809693738e-05,
1639
+ "loss": 0.0613,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 8.73134328358209,
1644
+ "grad_norm": 0.21864956617355347,
1645
+ "learning_rate": 9.877773673889701e-05,
1646
+ "loss": 0.0649,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 8.76865671641791,
1651
+ "grad_norm": 0.27523377537727356,
1652
+ "learning_rate": 9.8759502024387e-05,
1653
+ "loss": 0.0606,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 8.805970149253731,
1658
+ "grad_norm": 0.24805469810962677,
1659
+ "learning_rate": 9.87411340032603e-05,
1660
+ "loss": 0.0549,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 8.843283582089553,
1665
+ "grad_norm": 0.23070092499256134,
1666
+ "learning_rate": 9.872263272573443e-05,
1667
+ "loss": 0.0562,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 8.880597014925373,
1672
+ "grad_norm": 0.20833946764469147,
1673
+ "learning_rate": 9.870399824239117e-05,
1674
+ "loss": 0.05,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 8.917910447761194,
1679
+ "grad_norm": 0.34507372975349426,
1680
+ "learning_rate": 9.868523060417646e-05,
1681
+ "loss": 0.0613,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 8.955223880597014,
1686
+ "grad_norm": 0.32865110039711,
1687
+ "learning_rate": 9.86663298624003e-05,
1688
+ "loss": 0.0621,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 8.992537313432836,
1693
+ "grad_norm": 0.21305270493030548,
1694
+ "learning_rate": 9.864729606873663e-05,
1695
+ "loss": 0.0572,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 9.029850746268657,
1700
+ "grad_norm": 0.28193730115890503,
1701
+ "learning_rate": 9.862812927522309e-05,
1702
+ "loss": 0.0555,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 9.067164179104477,
1707
+ "grad_norm": 0.3953789472579956,
1708
+ "learning_rate": 9.860882953426099e-05,
1709
+ "loss": 0.0536,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 9.104477611940299,
1714
+ "grad_norm": 0.23013322055339813,
1715
+ "learning_rate": 9.858939689861506e-05,
1716
+ "loss": 0.0572,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 9.14179104477612,
1721
+ "grad_norm": 0.2906680107116699,
1722
+ "learning_rate": 9.856983142141339e-05,
1723
+ "loss": 0.0592,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 9.17910447761194,
1728
+ "grad_norm": 0.23490828275680542,
1729
+ "learning_rate": 9.855013315614725e-05,
1730
+ "loss": 0.0583,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 9.216417910447761,
1735
+ "grad_norm": 0.22825880348682404,
1736
+ "learning_rate": 9.853030215667093e-05,
1737
+ "loss": 0.059,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 9.253731343283581,
1742
+ "grad_norm": 0.25871285796165466,
1743
+ "learning_rate": 9.851033847720166e-05,
1744
+ "loss": 0.0555,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 9.291044776119403,
1749
+ "grad_norm": 0.27220776677131653,
1750
+ "learning_rate": 9.849024217231935e-05,
1751
+ "loss": 0.0542,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 9.328358208955224,
1756
+ "grad_norm": 0.26534005999565125,
1757
+ "learning_rate": 9.847001329696653e-05,
1758
+ "loss": 0.0526,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 9.365671641791044,
1763
+ "grad_norm": 0.33486032485961914,
1764
+ "learning_rate": 9.844965190644817e-05,
1765
+ "loss": 0.0563,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 9.402985074626866,
1770
+ "grad_norm": 0.2949483394622803,
1771
+ "learning_rate": 9.842915805643155e-05,
1772
+ "loss": 0.0556,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 9.440298507462687,
1777
+ "grad_norm": 0.24123981595039368,
1778
+ "learning_rate": 9.840853180294608e-05,
1779
+ "loss": 0.05,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 9.477611940298507,
1784
+ "grad_norm": 0.22536049783229828,
1785
+ "learning_rate": 9.838777320238312e-05,
1786
+ "loss": 0.0522,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 9.514925373134329,
1791
+ "grad_norm": 0.23206663131713867,
1792
+ "learning_rate": 9.836688231149592e-05,
1793
+ "loss": 0.0591,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 9.552238805970148,
1798
+ "grad_norm": 0.28573134541511536,
1799
+ "learning_rate": 9.834585918739936e-05,
1800
+ "loss": 0.0568,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 9.58955223880597,
1805
+ "grad_norm": 0.2628820538520813,
1806
+ "learning_rate": 9.832470388756987e-05,
1807
+ "loss": 0.0571,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 9.626865671641792,
1812
+ "grad_norm": 0.2880440652370453,
1813
+ "learning_rate": 9.830341646984521e-05,
1814
+ "loss": 0.0559,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 9.664179104477611,
1819
+ "grad_norm": 0.1786259263753891,
1820
+ "learning_rate": 9.82819969924244e-05,
1821
+ "loss": 0.058,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 9.701492537313433,
1826
+ "grad_norm": 0.3501608073711395,
1827
+ "learning_rate": 9.826044551386744e-05,
1828
+ "loss": 0.0523,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 9.738805970149254,
1833
+ "grad_norm": 0.24757252633571625,
1834
+ "learning_rate": 9.823876209309527e-05,
1835
+ "loss": 0.0587,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 9.776119402985074,
1840
+ "grad_norm": 0.2556290626525879,
1841
+ "learning_rate": 9.821694678938953e-05,
1842
+ "loss": 0.0555,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 9.813432835820896,
1847
+ "grad_norm": 0.2561217248439789,
1848
+ "learning_rate": 9.819499966239243e-05,
1849
+ "loss": 0.052,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 9.850746268656717,
1854
+ "grad_norm": 0.2776634097099304,
1855
+ "learning_rate": 9.817292077210659e-05,
1856
+ "loss": 0.0498,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 9.888059701492537,
1861
+ "grad_norm": 0.20668549835681915,
1862
+ "learning_rate": 9.815071017889482e-05,
1863
+ "loss": 0.0517,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 9.925373134328359,
1868
+ "grad_norm": 0.3100263178348541,
1869
+ "learning_rate": 9.812836794348004e-05,
1870
+ "loss": 0.0633,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 9.962686567164178,
1875
+ "grad_norm": 0.2780782878398895,
1876
+ "learning_rate": 9.81058941269451e-05,
1877
+ "loss": 0.0581,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 10.0,
1882
+ "grad_norm": 0.28903728723526,
1883
+ "learning_rate": 9.808328879073251e-05,
1884
+ "loss": 0.0538,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 10.037313432835822,
1889
+ "grad_norm": 0.22727562487125397,
1890
+ "learning_rate": 9.806055199664446e-05,
1891
+ "loss": 0.0491,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 10.074626865671641,
1896
+ "grad_norm": 0.267918199300766,
1897
+ "learning_rate": 9.803768380684242e-05,
1898
+ "loss": 0.0562,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 10.111940298507463,
1903
+ "grad_norm": 0.2988606095314026,
1904
+ "learning_rate": 9.801468428384716e-05,
1905
+ "loss": 0.0566,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 10.149253731343283,
1910
+ "grad_norm": 0.2710281312465668,
1911
+ "learning_rate": 9.799155349053851e-05,
1912
+ "loss": 0.0541,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 10.186567164179104,
1917
+ "grad_norm": 0.15320520102977753,
1918
+ "learning_rate": 9.796829149015517e-05,
1919
+ "loss": 0.0548,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 10.223880597014926,
1924
+ "grad_norm": 0.2653089463710785,
1925
+ "learning_rate": 9.794489834629455e-05,
1926
+ "loss": 0.0599,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 10.261194029850746,
1931
+ "grad_norm": 0.19223959743976593,
1932
+ "learning_rate": 9.792137412291265e-05,
1933
+ "loss": 0.0494,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 10.298507462686567,
1938
+ "grad_norm": 0.20455987751483917,
1939
+ "learning_rate": 9.789771888432375e-05,
1940
+ "loss": 0.0538,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 10.335820895522389,
1945
+ "grad_norm": 0.24908749759197235,
1946
+ "learning_rate": 9.787393269520039e-05,
1947
+ "loss": 0.0481,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 10.373134328358208,
1952
+ "grad_norm": 0.3131813406944275,
1953
+ "learning_rate": 9.785001562057309e-05,
1954
+ "loss": 0.0526,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 10.41044776119403,
1959
+ "grad_norm": 0.24828971922397614,
1960
+ "learning_rate": 9.782596772583026e-05,
1961
+ "loss": 0.0489,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 10.447761194029852,
1966
+ "grad_norm": 0.21727119386196136,
1967
+ "learning_rate": 9.780178907671789e-05,
1968
+ "loss": 0.0532,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 10.485074626865671,
1973
+ "grad_norm": 0.20279547572135925,
1974
+ "learning_rate": 9.777747973933948e-05,
1975
+ "loss": 0.0565,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 10.522388059701493,
1980
+ "grad_norm": 0.17726702988147736,
1981
+ "learning_rate": 9.775303978015585e-05,
1982
+ "loss": 0.0437,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 10.559701492537313,
1987
+ "grad_norm": 0.18961119651794434,
1988
+ "learning_rate": 9.772846926598491e-05,
1989
+ "loss": 0.0584,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 10.597014925373134,
1994
+ "grad_norm": 0.2498980015516281,
1995
+ "learning_rate": 9.77037682640015e-05,
1996
+ "loss": 0.0496,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 10.634328358208956,
2001
+ "grad_norm": 0.16978798806667328,
2002
+ "learning_rate": 9.767893684173721e-05,
2003
+ "loss": 0.0469,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 10.671641791044776,
2008
+ "grad_norm": 0.16128584742546082,
2009
+ "learning_rate": 9.765397506708023e-05,
2010
+ "loss": 0.0533,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 10.708955223880597,
2015
+ "grad_norm": 0.20463155210018158,
2016
+ "learning_rate": 9.762888300827507e-05,
2017
+ "loss": 0.0464,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 10.746268656716419,
2022
+ "grad_norm": 0.30601629614830017,
2023
+ "learning_rate": 9.760366073392246e-05,
2024
+ "loss": 0.0489,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 10.783582089552239,
2029
+ "grad_norm": 0.2730671763420105,
2030
+ "learning_rate": 9.757830831297914e-05,
2031
+ "loss": 0.0495,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 10.82089552238806,
2036
+ "grad_norm": 0.251432865858078,
2037
+ "learning_rate": 9.755282581475769e-05,
2038
+ "loss": 0.0549,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 10.85820895522388,
2043
+ "grad_norm": 0.26670166850090027,
2044
+ "learning_rate": 9.752721330892624e-05,
2045
+ "loss": 0.061,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 10.895522388059701,
2050
+ "grad_norm": 0.2965967655181885,
2051
+ "learning_rate": 9.750147086550844e-05,
2052
+ "loss": 0.0473,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 10.932835820895523,
2057
+ "grad_norm": 0.683840274810791,
2058
+ "learning_rate": 9.747559855488313e-05,
2059
+ "loss": 0.0509,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 10.970149253731343,
2064
+ "grad_norm": 0.25740495324134827,
2065
+ "learning_rate": 9.744959644778422e-05,
2066
+ "loss": 0.0515,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 11.007462686567164,
2071
+ "grad_norm": 0.2880542278289795,
2072
+ "learning_rate": 9.742346461530048e-05,
2073
+ "loss": 0.0482,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 11.044776119402986,
2078
+ "grad_norm": 0.45032551884651184,
2079
+ "learning_rate": 9.739720312887535e-05,
2080
+ "loss": 0.0557,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 11.082089552238806,
2085
+ "grad_norm": 0.2829900085926056,
2086
+ "learning_rate": 9.73708120603067e-05,
2087
+ "loss": 0.052,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 11.119402985074627,
2092
+ "grad_norm": 0.309597373008728,
2093
+ "learning_rate": 9.734429148174675e-05,
2094
+ "loss": 0.0541,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 11.156716417910447,
2099
+ "grad_norm": 0.2433389127254486,
2100
+ "learning_rate": 9.731764146570173e-05,
2101
+ "loss": 0.0482,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 11.194029850746269,
2106
+ "grad_norm": 0.24458132684230804,
2107
+ "learning_rate": 9.729086208503174e-05,
2108
+ "loss": 0.0505,
2109
+ "step": 3000
2110
+ },
2111
+ {
2112
+ "epoch": 11.23134328358209,
2113
+ "grad_norm": 0.2305087298154831,
2114
+ "learning_rate": 9.726395341295062e-05,
2115
+ "loss": 0.0504,
2116
+ "step": 3010
2117
+ },
2118
+ {
2119
+ "epoch": 11.26865671641791,
2120
+ "grad_norm": 0.18110457062721252,
2121
+ "learning_rate": 9.723691552302562e-05,
2122
+ "loss": 0.0575,
2123
+ "step": 3020
2124
+ },
2125
+ {
2126
+ "epoch": 11.305970149253731,
2127
+ "grad_norm": 0.20407621562480927,
2128
+ "learning_rate": 9.720974848917735e-05,
2129
+ "loss": 0.0494,
2130
+ "step": 3030
2131
+ },
2132
+ {
2133
+ "epoch": 11.343283582089553,
2134
+ "grad_norm": 0.25924697518348694,
2135
+ "learning_rate": 9.718245238567939e-05,
2136
+ "loss": 0.0472,
2137
+ "step": 3040
2138
+ },
2139
+ {
2140
+ "epoch": 11.380597014925373,
2141
+ "grad_norm": 0.23041822016239166,
2142
+ "learning_rate": 9.715502728715826e-05,
2143
+ "loss": 0.0481,
2144
+ "step": 3050
2145
+ },
2146
+ {
2147
+ "epoch": 11.417910447761194,
2148
+ "grad_norm": 0.25381171703338623,
2149
+ "learning_rate": 9.712747326859315e-05,
2150
+ "loss": 0.0543,
2151
+ "step": 3060
2152
+ },
2153
+ {
2154
+ "epoch": 11.455223880597014,
2155
+ "grad_norm": 0.18027640879154205,
2156
+ "learning_rate": 9.709979040531569e-05,
2157
+ "loss": 0.055,
2158
+ "step": 3070
2159
+ },
2160
+ {
2161
+ "epoch": 11.492537313432836,
2162
+ "grad_norm": 0.2954868674278259,
2163
+ "learning_rate": 9.707197877300974e-05,
2164
+ "loss": 0.0473,
2165
+ "step": 3080
2166
+ },
2167
+ {
2168
+ "epoch": 11.529850746268657,
2169
+ "grad_norm": 0.25323861837387085,
2170
+ "learning_rate": 9.704403844771128e-05,
2171
+ "loss": 0.0509,
2172
+ "step": 3090
2173
+ },
2174
+ {
2175
+ "epoch": 11.567164179104477,
2176
+ "grad_norm": 0.36910176277160645,
2177
+ "learning_rate": 9.701596950580806e-05,
2178
+ "loss": 0.0504,
2179
+ "step": 3100
2180
+ },
2181
+ {
2182
+ "epoch": 11.604477611940299,
2183
+ "grad_norm": 0.34199246764183044,
2184
+ "learning_rate": 9.698777202403953e-05,
2185
+ "loss": 0.0526,
2186
+ "step": 3110
2187
+ },
2188
+ {
2189
+ "epoch": 11.64179104477612,
2190
+ "grad_norm": 0.2146557718515396,
2191
+ "learning_rate": 9.695944607949649e-05,
2192
+ "loss": 0.0579,
2193
+ "step": 3120
2194
+ },
2195
+ {
2196
+ "epoch": 11.67910447761194,
2197
+ "grad_norm": 0.20559175312519073,
2198
+ "learning_rate": 9.693099174962103e-05,
2199
+ "loss": 0.0514,
2200
+ "step": 3130
2201
+ },
2202
+ {
2203
+ "epoch": 11.716417910447761,
2204
+ "grad_norm": 0.2689419090747833,
2205
+ "learning_rate": 9.690240911220618e-05,
2206
+ "loss": 0.0534,
2207
+ "step": 3140
2208
+ },
2209
+ {
2210
+ "epoch": 11.753731343283581,
2211
+ "grad_norm": 0.34870603680610657,
2212
+ "learning_rate": 9.687369824539577e-05,
2213
+ "loss": 0.0485,
2214
+ "step": 3150
2215
+ },
2216
+ {
2217
+ "epoch": 11.791044776119403,
2218
+ "grad_norm": 0.15433363616466522,
2219
+ "learning_rate": 9.684485922768422e-05,
2220
+ "loss": 0.0418,
2221
+ "step": 3160
2222
+ },
2223
+ {
2224
+ "epoch": 11.828358208955224,
2225
+ "grad_norm": 0.26874423027038574,
2226
+ "learning_rate": 9.681589213791633e-05,
2227
+ "loss": 0.0537,
2228
+ "step": 3170
2229
+ },
2230
+ {
2231
+ "epoch": 11.865671641791044,
2232
+ "grad_norm": 0.3361654281616211,
2233
+ "learning_rate": 9.6786797055287e-05,
2234
+ "loss": 0.0474,
2235
+ "step": 3180
2236
+ },
2237
+ {
2238
+ "epoch": 11.902985074626866,
2239
+ "grad_norm": 0.17938771843910217,
2240
+ "learning_rate": 9.675757405934103e-05,
2241
+ "loss": 0.0443,
2242
+ "step": 3190
2243
+ },
2244
+ {
2245
+ "epoch": 11.940298507462687,
2246
+ "grad_norm": 0.31368622183799744,
2247
+ "learning_rate": 9.672822322997305e-05,
2248
+ "loss": 0.0594,
2249
+ "step": 3200
2250
+ },
2251
+ {
2252
+ "epoch": 11.977611940298507,
2253
+ "grad_norm": 0.16268151998519897,
2254
+ "learning_rate": 9.669874464742705e-05,
2255
+ "loss": 0.0487,
2256
+ "step": 3210
2257
+ },
2258
+ {
2259
+ "epoch": 12.014925373134329,
2260
+ "grad_norm": 0.23879969120025635,
2261
+ "learning_rate": 9.66691383922964e-05,
2262
+ "loss": 0.0484,
2263
+ "step": 3220
2264
+ },
2265
+ {
2266
+ "epoch": 12.052238805970148,
2267
+ "grad_norm": 0.2321789413690567,
2268
+ "learning_rate": 9.663940454552342e-05,
2269
+ "loss": 0.051,
2270
+ "step": 3230
2271
+ },
2272
+ {
2273
+ "epoch": 12.08955223880597,
2274
+ "grad_norm": 0.22873088717460632,
2275
+ "learning_rate": 9.660954318839933e-05,
2276
+ "loss": 0.0406,
2277
+ "step": 3240
2278
+ },
2279
+ {
2280
+ "epoch": 12.126865671641792,
2281
+ "grad_norm": 0.3767557740211487,
2282
+ "learning_rate": 9.657955440256395e-05,
2283
+ "loss": 0.0432,
2284
+ "step": 3250
2285
+ },
2286
+ {
2287
+ "epoch": 12.164179104477611,
2288
+ "grad_norm": 0.21569453179836273,
2289
+ "learning_rate": 9.654943827000548e-05,
2290
+ "loss": 0.0528,
2291
+ "step": 3260
2292
+ },
2293
+ {
2294
+ "epoch": 12.201492537313433,
2295
+ "grad_norm": 0.23698291182518005,
2296
+ "learning_rate": 9.651919487306025e-05,
2297
+ "loss": 0.0457,
2298
+ "step": 3270
2299
+ },
2300
+ {
2301
+ "epoch": 12.238805970149254,
2302
+ "grad_norm": 0.21086478233337402,
2303
+ "learning_rate": 9.648882429441257e-05,
2304
+ "loss": 0.0508,
2305
+ "step": 3280
2306
+ },
2307
+ {
2308
+ "epoch": 12.276119402985074,
2309
+ "grad_norm": 0.19763463735580444,
2310
+ "learning_rate": 9.645832661709444e-05,
2311
+ "loss": 0.0497,
2312
+ "step": 3290
2313
+ },
2314
+ {
2315
+ "epoch": 12.313432835820896,
2316
+ "grad_norm": 0.18413852155208588,
2317
+ "learning_rate": 9.642770192448536e-05,
2318
+ "loss": 0.0441,
2319
+ "step": 3300
2320
+ },
2321
+ {
2322
+ "epoch": 12.350746268656717,
2323
+ "grad_norm": 0.13946911692619324,
2324
+ "learning_rate": 9.639695030031204e-05,
2325
+ "loss": 0.0453,
2326
+ "step": 3310
2327
+ },
2328
+ {
2329
+ "epoch": 12.388059701492537,
2330
+ "grad_norm": 0.21613670885562897,
2331
+ "learning_rate": 9.636607182864827e-05,
2332
+ "loss": 0.0511,
2333
+ "step": 3320
2334
+ },
2335
+ {
2336
+ "epoch": 12.425373134328359,
2337
+ "grad_norm": 0.24953646957874298,
2338
+ "learning_rate": 9.63350665939146e-05,
2339
+ "loss": 0.0451,
2340
+ "step": 3330
2341
+ },
2342
+ {
2343
+ "epoch": 12.462686567164178,
2344
+ "grad_norm": 0.2993795871734619,
2345
+ "learning_rate": 9.630393468087818e-05,
2346
+ "loss": 0.0469,
2347
+ "step": 3340
2348
+ },
2349
+ {
2350
+ "epoch": 12.5,
2351
+ "grad_norm": 0.2261819839477539,
2352
+ "learning_rate": 9.627267617465243e-05,
2353
+ "loss": 0.0484,
2354
+ "step": 3350
2355
+ },
2356
+ {
2357
+ "epoch": 12.537313432835822,
2358
+ "grad_norm": 0.23026186227798462,
2359
+ "learning_rate": 9.624129116069694e-05,
2360
+ "loss": 0.0452,
2361
+ "step": 3360
2362
+ },
2363
+ {
2364
+ "epoch": 12.574626865671641,
2365
+ "grad_norm": 0.27859947085380554,
2366
+ "learning_rate": 9.620977972481716e-05,
2367
+ "loss": 0.0593,
2368
+ "step": 3370
2369
+ },
2370
+ {
2371
+ "epoch": 12.611940298507463,
2372
+ "grad_norm": 0.23060785233974457,
2373
+ "learning_rate": 9.617814195316411e-05,
2374
+ "loss": 0.05,
2375
+ "step": 3380
2376
+ },
2377
+ {
2378
+ "epoch": 12.649253731343283,
2379
+ "grad_norm": 0.20185025036334991,
2380
+ "learning_rate": 9.614637793223425e-05,
2381
+ "loss": 0.0573,
2382
+ "step": 3390
2383
+ },
2384
+ {
2385
+ "epoch": 12.686567164179104,
2386
+ "grad_norm": 0.3584498167037964,
2387
+ "learning_rate": 9.611448774886924e-05,
2388
+ "loss": 0.052,
2389
+ "step": 3400
2390
+ },
2391
+ {
2392
+ "epoch": 12.723880597014926,
2393
+ "grad_norm": 0.19336827099323273,
2394
+ "learning_rate": 9.60824714902556e-05,
2395
+ "loss": 0.0535,
2396
+ "step": 3410
2397
+ },
2398
+ {
2399
+ "epoch": 12.761194029850746,
2400
+ "grad_norm": 0.22223635017871857,
2401
+ "learning_rate": 9.605032924392457e-05,
2402
+ "loss": 0.05,
2403
+ "step": 3420
2404
+ },
2405
+ {
2406
+ "epoch": 12.798507462686567,
2407
+ "grad_norm": 0.17108851671218872,
2408
+ "learning_rate": 9.601806109775179e-05,
2409
+ "loss": 0.0475,
2410
+ "step": 3430
2411
+ },
2412
+ {
2413
+ "epoch": 12.835820895522389,
2414
+ "grad_norm": 0.3861902952194214,
2415
+ "learning_rate": 9.598566713995718e-05,
2416
+ "loss": 0.0439,
2417
+ "step": 3440
2418
+ },
2419
+ {
2420
+ "epoch": 12.873134328358208,
2421
+ "grad_norm": 0.18927253782749176,
2422
+ "learning_rate": 9.595314745910456e-05,
2423
+ "loss": 0.052,
2424
+ "step": 3450
2425
+ },
2426
+ {
2427
+ "epoch": 12.91044776119403,
2428
+ "grad_norm": 0.21963383257389069,
2429
+ "learning_rate": 9.59205021441015e-05,
2430
+ "loss": 0.0504,
2431
+ "step": 3460
2432
+ },
2433
+ {
2434
+ "epoch": 12.947761194029852,
2435
+ "grad_norm": 0.18016670644283295,
2436
+ "learning_rate": 9.588773128419906e-05,
2437
+ "loss": 0.0467,
2438
+ "step": 3470
2439
+ },
2440
+ {
2441
+ "epoch": 12.985074626865671,
2442
+ "grad_norm": 0.1776365041732788,
2443
+ "learning_rate": 9.58548349689915e-05,
2444
+ "loss": 0.0414,
2445
+ "step": 3480
2446
+ },
2447
+ {
2448
+ "epoch": 13.022388059701493,
2449
+ "grad_norm": 0.2616482973098755,
2450
+ "learning_rate": 9.582181328841611e-05,
2451
+ "loss": 0.0442,
2452
+ "step": 3490
2453
+ },
2454
+ {
2455
+ "epoch": 13.059701492537313,
2456
+ "grad_norm": 0.20341171324253082,
2457
+ "learning_rate": 9.578866633275288e-05,
2458
+ "loss": 0.0533,
2459
+ "step": 3500
2460
+ },
2461
+ {
2462
+ "epoch": 13.097014925373134,
2463
+ "grad_norm": 0.2223699688911438,
2464
+ "learning_rate": 9.575539419262434e-05,
2465
+ "loss": 0.0458,
2466
+ "step": 3510
2467
+ },
2468
+ {
2469
+ "epoch": 13.134328358208956,
2470
+ "grad_norm": 0.22557464241981506,
2471
+ "learning_rate": 9.572199695899522e-05,
2472
+ "loss": 0.0445,
2473
+ "step": 3520
2474
+ },
2475
+ {
2476
+ "epoch": 13.171641791044776,
2477
+ "grad_norm": 0.25104308128356934,
2478
+ "learning_rate": 9.568847472317232e-05,
2479
+ "loss": 0.0435,
2480
+ "step": 3530
2481
+ },
2482
+ {
2483
+ "epoch": 13.208955223880597,
2484
+ "grad_norm": 0.18720711767673492,
2485
+ "learning_rate": 9.565482757680415e-05,
2486
+ "loss": 0.0453,
2487
+ "step": 3540
2488
+ },
2489
+ {
2490
+ "epoch": 13.246268656716419,
2491
+ "grad_norm": 0.16838951408863068,
2492
+ "learning_rate": 9.562105561188069e-05,
2493
+ "loss": 0.0505,
2494
+ "step": 3550
2495
+ },
2496
+ {
2497
+ "epoch": 13.283582089552239,
2498
+ "grad_norm": 0.31681734323501587,
2499
+ "learning_rate": 9.558715892073323e-05,
2500
+ "loss": 0.0494,
2501
+ "step": 3560
2502
+ },
2503
+ {
2504
+ "epoch": 13.32089552238806,
2505
+ "grad_norm": 0.2390700727701187,
2506
+ "learning_rate": 9.555313759603402e-05,
2507
+ "loss": 0.0538,
2508
+ "step": 3570
2509
+ },
2510
+ {
2511
+ "epoch": 13.35820895522388,
2512
+ "grad_norm": 0.20680709183216095,
2513
+ "learning_rate": 9.551899173079607e-05,
2514
+ "loss": 0.0519,
2515
+ "step": 3580
2516
+ },
2517
+ {
2518
+ "epoch": 13.395522388059701,
2519
+ "grad_norm": 0.2758580148220062,
2520
+ "learning_rate": 9.548472141837286e-05,
2521
+ "loss": 0.0512,
2522
+ "step": 3590
2523
+ },
2524
+ {
2525
+ "epoch": 13.432835820895523,
2526
+ "grad_norm": 0.3653097450733185,
2527
+ "learning_rate": 9.545032675245813e-05,
2528
+ "loss": 0.0496,
2529
+ "step": 3600
2530
+ },
2531
+ {
2532
+ "epoch": 13.470149253731343,
2533
+ "grad_norm": 0.23886866867542267,
2534
+ "learning_rate": 9.541580782708557e-05,
2535
+ "loss": 0.0455,
2536
+ "step": 3610
2537
+ },
2538
+ {
2539
+ "epoch": 13.507462686567164,
2540
+ "grad_norm": 0.3280908465385437,
2541
+ "learning_rate": 9.538116473662861e-05,
2542
+ "loss": 0.0489,
2543
+ "step": 3620
2544
+ },
2545
+ {
2546
+ "epoch": 13.544776119402986,
2547
+ "grad_norm": 0.20268180966377258,
2548
+ "learning_rate": 9.534639757580013e-05,
2549
+ "loss": 0.0484,
2550
+ "step": 3630
2551
+ },
2552
+ {
2553
+ "epoch": 13.582089552238806,
2554
+ "grad_norm": 0.2582015097141266,
2555
+ "learning_rate": 9.531150643965223e-05,
2556
+ "loss": 0.0487,
2557
+ "step": 3640
2558
+ },
2559
+ {
2560
+ "epoch": 13.619402985074627,
2561
+ "grad_norm": 0.18157973885536194,
2562
+ "learning_rate": 9.527649142357596e-05,
2563
+ "loss": 0.0496,
2564
+ "step": 3650
2565
+ },
2566
+ {
2567
+ "epoch": 13.656716417910447,
2568
+ "grad_norm": 0.22841542959213257,
2569
+ "learning_rate": 9.524135262330098e-05,
2570
+ "loss": 0.0467,
2571
+ "step": 3660
2572
+ },
2573
+ {
2574
+ "epoch": 13.694029850746269,
2575
+ "grad_norm": 0.2519935369491577,
2576
+ "learning_rate": 9.520609013489547e-05,
2577
+ "loss": 0.0487,
2578
+ "step": 3670
2579
+ },
2580
+ {
2581
+ "epoch": 13.73134328358209,
2582
+ "grad_norm": 0.24680495262145996,
2583
+ "learning_rate": 9.517070405476575e-05,
2584
+ "loss": 0.0457,
2585
+ "step": 3680
2586
+ },
2587
+ {
2588
+ "epoch": 13.76865671641791,
2589
+ "grad_norm": 0.26362067461013794,
2590
+ "learning_rate": 9.513519447965595e-05,
2591
+ "loss": 0.0495,
2592
+ "step": 3690
2593
+ },
2594
+ {
2595
+ "epoch": 13.805970149253731,
2596
+ "grad_norm": 0.3240712583065033,
2597
+ "learning_rate": 9.509956150664796e-05,
2598
+ "loss": 0.0496,
2599
+ "step": 3700
2600
+ },
2601
+ {
2602
+ "epoch": 13.843283582089553,
2603
+ "grad_norm": 0.21009013056755066,
2604
+ "learning_rate": 9.50638052331609e-05,
2605
+ "loss": 0.0457,
2606
+ "step": 3710
2607
+ },
2608
+ {
2609
+ "epoch": 13.880597014925373,
2610
+ "grad_norm": 0.1669154316186905,
2611
+ "learning_rate": 9.502792575695112e-05,
2612
+ "loss": 0.0496,
2613
+ "step": 3720
2614
+ },
2615
+ {
2616
+ "epoch": 13.917910447761194,
2617
+ "grad_norm": 0.22347605228424072,
2618
+ "learning_rate": 9.499192317611167e-05,
2619
+ "loss": 0.0426,
2620
+ "step": 3730
2621
+ },
2622
+ {
2623
+ "epoch": 13.955223880597014,
2624
+ "grad_norm": 0.15208907425403595,
2625
+ "learning_rate": 9.49557975890723e-05,
2626
+ "loss": 0.0447,
2627
+ "step": 3740
2628
+ },
2629
+ {
2630
+ "epoch": 13.992537313432836,
2631
+ "grad_norm": 0.3206101059913635,
2632
+ "learning_rate": 9.491954909459895e-05,
2633
+ "loss": 0.0471,
2634
+ "step": 3750
2635
+ },
2636
+ {
2637
+ "epoch": 14.029850746268657,
2638
+ "grad_norm": 0.15873713791370392,
2639
+ "learning_rate": 9.488317779179361e-05,
2640
+ "loss": 0.0401,
2641
+ "step": 3760
2642
+ },
2643
+ {
2644
+ "epoch": 14.067164179104477,
2645
+ "grad_norm": 0.19690357148647308,
2646
+ "learning_rate": 9.484668378009408e-05,
2647
+ "loss": 0.0491,
2648
+ "step": 3770
2649
+ },
2650
+ {
2651
+ "epoch": 14.104477611940299,
2652
+ "grad_norm": 0.3211113214492798,
2653
+ "learning_rate": 9.481006715927351e-05,
2654
+ "loss": 0.049,
2655
+ "step": 3780
2656
+ },
2657
+ {
2658
+ "epoch": 14.14179104477612,
2659
+ "grad_norm": 0.27657604217529297,
2660
+ "learning_rate": 9.477332802944044e-05,
2661
+ "loss": 0.0396,
2662
+ "step": 3790
2663
+ },
2664
+ {
2665
+ "epoch": 14.17910447761194,
2666
+ "grad_norm": 0.20194031298160553,
2667
+ "learning_rate": 9.473646649103818e-05,
2668
+ "loss": 0.0442,
2669
+ "step": 3800
2670
+ },
2671
+ {
2672
+ "epoch": 14.216417910447761,
2673
+ "grad_norm": 0.20344595611095428,
2674
+ "learning_rate": 9.46994826448448e-05,
2675
+ "loss": 0.0427,
2676
+ "step": 3810
2677
+ },
2678
+ {
2679
+ "epoch": 14.253731343283581,
2680
+ "grad_norm": 0.2067718505859375,
2681
+ "learning_rate": 9.46623765919727e-05,
2682
+ "loss": 0.0501,
2683
+ "step": 3820
2684
+ },
2685
+ {
2686
+ "epoch": 14.291044776119403,
2687
+ "grad_norm": 0.29719170928001404,
2688
+ "learning_rate": 9.462514843386845e-05,
2689
+ "loss": 0.0519,
2690
+ "step": 3830
2691
+ },
2692
+ {
2693
+ "epoch": 14.328358208955224,
2694
+ "grad_norm": 0.2347182184457779,
2695
+ "learning_rate": 9.458779827231237e-05,
2696
+ "loss": 0.0413,
2697
+ "step": 3840
2698
+ },
2699
+ {
2700
+ "epoch": 14.365671641791044,
2701
+ "grad_norm": 0.1558852344751358,
2702
+ "learning_rate": 9.45503262094184e-05,
2703
+ "loss": 0.0442,
2704
+ "step": 3850
2705
+ },
2706
+ {
2707
+ "epoch": 14.402985074626866,
2708
+ "grad_norm": 0.23085005581378937,
2709
+ "learning_rate": 9.451273234763371e-05,
2710
+ "loss": 0.047,
2711
+ "step": 3860
2712
+ },
2713
+ {
2714
+ "epoch": 14.440298507462687,
2715
+ "grad_norm": 0.1515151560306549,
2716
+ "learning_rate": 9.447501678973852e-05,
2717
+ "loss": 0.0481,
2718
+ "step": 3870
2719
+ },
2720
+ {
2721
+ "epoch": 14.477611940298507,
2722
+ "grad_norm": 0.1916729211807251,
2723
+ "learning_rate": 9.443717963884569e-05,
2724
+ "loss": 0.0474,
2725
+ "step": 3880
2726
+ },
2727
+ {
2728
+ "epoch": 14.514925373134329,
2729
+ "grad_norm": 0.2536492943763733,
2730
+ "learning_rate": 9.439922099840054e-05,
2731
+ "loss": 0.0382,
2732
+ "step": 3890
2733
+ },
2734
+ {
2735
+ "epoch": 14.552238805970148,
2736
+ "grad_norm": 0.1672086864709854,
2737
+ "learning_rate": 9.43611409721806e-05,
2738
+ "loss": 0.0497,
2739
+ "step": 3900
2740
+ },
2741
+ {
2742
+ "epoch": 14.58955223880597,
2743
+ "grad_norm": 0.3644237518310547,
2744
+ "learning_rate": 9.432293966429514e-05,
2745
+ "loss": 0.0444,
2746
+ "step": 3910
2747
+ },
2748
+ {
2749
+ "epoch": 14.626865671641792,
2750
+ "grad_norm": 0.20307251811027527,
2751
+ "learning_rate": 9.428461717918511e-05,
2752
+ "loss": 0.0452,
2753
+ "step": 3920
2754
+ },
2755
+ {
2756
+ "epoch": 14.664179104477611,
2757
+ "grad_norm": 0.20441733300685883,
2758
+ "learning_rate": 9.424617362162271e-05,
2759
+ "loss": 0.0454,
2760
+ "step": 3930
2761
+ },
2762
+ {
2763
+ "epoch": 14.701492537313433,
2764
+ "grad_norm": 0.26315611600875854,
2765
+ "learning_rate": 9.420760909671118e-05,
2766
+ "loss": 0.0486,
2767
+ "step": 3940
2768
+ },
2769
+ {
2770
+ "epoch": 14.738805970149254,
2771
+ "grad_norm": 0.1983092874288559,
2772
+ "learning_rate": 9.416892370988444e-05,
2773
+ "loss": 0.0483,
2774
+ "step": 3950
2775
+ },
2776
+ {
2777
+ "epoch": 14.776119402985074,
2778
+ "grad_norm": 0.18301443755626678,
2779
+ "learning_rate": 9.413011756690685e-05,
2780
+ "loss": 0.0456,
2781
+ "step": 3960
2782
+ },
2783
+ {
2784
+ "epoch": 14.813432835820896,
2785
+ "grad_norm": 0.2433597594499588,
2786
+ "learning_rate": 9.409119077387294e-05,
2787
+ "loss": 0.0463,
2788
+ "step": 3970
2789
+ },
2790
+ {
2791
+ "epoch": 14.850746268656717,
2792
+ "grad_norm": 0.27949392795562744,
2793
+ "learning_rate": 9.405214343720707e-05,
2794
+ "loss": 0.0412,
2795
+ "step": 3980
2796
+ },
2797
+ {
2798
+ "epoch": 14.888059701492537,
2799
+ "grad_norm": 0.22806599736213684,
2800
+ "learning_rate": 9.401297566366318e-05,
2801
+ "loss": 0.0448,
2802
+ "step": 3990
2803
+ },
2804
+ {
2805
+ "epoch": 14.925373134328359,
2806
+ "grad_norm": 0.25421562790870667,
2807
+ "learning_rate": 9.397368756032445e-05,
2808
+ "loss": 0.0426,
2809
+ "step": 4000
2810
+ },
2811
+ {
2812
+ "epoch": 14.962686567164178,
2813
+ "grad_norm": 0.2436474859714508,
2814
+ "learning_rate": 9.393427923460308e-05,
2815
+ "loss": 0.0474,
2816
+ "step": 4010
2817
+ },
2818
+ {
2819
+ "epoch": 15.0,
2820
+ "grad_norm": 0.3756405711174011,
2821
+ "learning_rate": 9.389475079423988e-05,
2822
+ "loss": 0.0438,
2823
+ "step": 4020
2824
+ },
2825
+ {
2826
+ "epoch": 15.037313432835822,
2827
+ "grad_norm": 0.25687697529792786,
2828
+ "learning_rate": 9.385510234730415e-05,
2829
+ "loss": 0.0435,
2830
+ "step": 4030
2831
+ },
2832
+ {
2833
+ "epoch": 15.074626865671641,
2834
+ "grad_norm": 0.17263716459274292,
2835
+ "learning_rate": 9.381533400219318e-05,
2836
+ "loss": 0.0455,
2837
+ "step": 4040
2838
+ },
2839
+ {
2840
+ "epoch": 15.111940298507463,
2841
+ "grad_norm": 0.2471216470003128,
2842
+ "learning_rate": 9.377544586763215e-05,
2843
+ "loss": 0.0429,
2844
+ "step": 4050
2845
+ },
2846
+ {
2847
+ "epoch": 15.149253731343283,
2848
+ "grad_norm": 0.20195460319519043,
2849
+ "learning_rate": 9.373543805267368e-05,
2850
+ "loss": 0.0432,
2851
+ "step": 4060
2852
+ },
2853
+ {
2854
+ "epoch": 15.186567164179104,
2855
+ "grad_norm": 0.1709851622581482,
2856
+ "learning_rate": 9.369531066669758e-05,
2857
+ "loss": 0.0477,
2858
+ "step": 4070
2859
+ },
2860
+ {
2861
+ "epoch": 15.223880597014926,
2862
+ "grad_norm": 0.23063932359218597,
2863
+ "learning_rate": 9.365506381941066e-05,
2864
+ "loss": 0.0379,
2865
+ "step": 4080
2866
+ },
2867
+ {
2868
+ "epoch": 15.261194029850746,
2869
+ "grad_norm": 0.3265426754951477,
2870
+ "learning_rate": 9.36146976208462e-05,
2871
+ "loss": 0.0435,
2872
+ "step": 4090
2873
+ },
2874
+ {
2875
+ "epoch": 15.298507462686567,
2876
+ "grad_norm": 0.26373934745788574,
2877
+ "learning_rate": 9.357421218136386e-05,
2878
+ "loss": 0.047,
2879
+ "step": 4100
2880
+ },
2881
+ {
2882
+ "epoch": 15.335820895522389,
2883
+ "grad_norm": 0.16861388087272644,
2884
+ "learning_rate": 9.353360761164931e-05,
2885
+ "loss": 0.0448,
2886
+ "step": 4110
2887
+ },
2888
+ {
2889
+ "epoch": 15.373134328358208,
2890
+ "grad_norm": 0.303790807723999,
2891
+ "learning_rate": 9.349288402271388e-05,
2892
+ "loss": 0.0396,
2893
+ "step": 4120
2894
+ },
2895
+ {
2896
+ "epoch": 15.41044776119403,
2897
+ "grad_norm": 0.1940719038248062,
2898
+ "learning_rate": 9.345204152589428e-05,
2899
+ "loss": 0.0474,
2900
+ "step": 4130
2901
+ },
2902
+ {
2903
+ "epoch": 15.447761194029852,
2904
+ "grad_norm": 0.34091615676879883,
2905
+ "learning_rate": 9.341108023285238e-05,
2906
+ "loss": 0.0424,
2907
+ "step": 4140
2908
+ },
2909
+ {
2910
+ "epoch": 15.485074626865671,
2911
+ "grad_norm": 0.27036693692207336,
2912
+ "learning_rate": 9.337000025557476e-05,
2913
+ "loss": 0.0482,
2914
+ "step": 4150
2915
+ },
2916
+ {
2917
+ "epoch": 15.522388059701493,
2918
+ "grad_norm": 0.16908007860183716,
2919
+ "learning_rate": 9.332880170637252e-05,
2920
+ "loss": 0.0381,
2921
+ "step": 4160
2922
+ },
2923
+ {
2924
+ "epoch": 15.559701492537313,
2925
+ "grad_norm": 0.23332923650741577,
2926
+ "learning_rate": 9.328748469788093e-05,
2927
+ "loss": 0.0427,
2928
+ "step": 4170
2929
+ },
2930
+ {
2931
+ "epoch": 15.597014925373134,
2932
+ "grad_norm": 0.16899706423282623,
2933
+ "learning_rate": 9.32460493430591e-05,
2934
+ "loss": 0.0439,
2935
+ "step": 4180
2936
+ },
2937
+ {
2938
+ "epoch": 15.634328358208956,
2939
+ "grad_norm": 0.12869524955749512,
2940
+ "learning_rate": 9.320449575518972e-05,
2941
+ "loss": 0.0481,
2942
+ "step": 4190
2943
+ },
2944
+ {
2945
+ "epoch": 15.671641791044776,
2946
+ "grad_norm": 0.21159130334854126,
2947
+ "learning_rate": 9.316282404787871e-05,
2948
+ "loss": 0.0446,
2949
+ "step": 4200
2950
+ },
2951
+ {
2952
+ "epoch": 15.708955223880597,
2953
+ "grad_norm": 0.1849961131811142,
2954
+ "learning_rate": 9.31210343350549e-05,
2955
+ "loss": 0.041,
2956
+ "step": 4210
2957
+ },
2958
+ {
2959
+ "epoch": 15.746268656716419,
2960
+ "grad_norm": 0.16107840836048126,
2961
+ "learning_rate": 9.30791267309698e-05,
2962
+ "loss": 0.0429,
2963
+ "step": 4220
2964
+ },
2965
+ {
2966
+ "epoch": 15.783582089552239,
2967
+ "grad_norm": 0.14206446707248688,
2968
+ "learning_rate": 9.30371013501972e-05,
2969
+ "loss": 0.0409,
2970
+ "step": 4230
2971
+ },
2972
+ {
2973
+ "epoch": 15.82089552238806,
2974
+ "grad_norm": 0.2168441116809845,
2975
+ "learning_rate": 9.299495830763286e-05,
2976
+ "loss": 0.0413,
2977
+ "step": 4240
2978
+ },
2979
+ {
2980
+ "epoch": 15.85820895522388,
2981
+ "grad_norm": 0.21431951224803925,
2982
+ "learning_rate": 9.295269771849427e-05,
2983
+ "loss": 0.0472,
2984
+ "step": 4250
2985
+ },
2986
+ {
2987
+ "epoch": 15.895522388059701,
2988
+ "grad_norm": 0.16851255297660828,
2989
+ "learning_rate": 9.291031969832026e-05,
2990
+ "loss": 0.0508,
2991
+ "step": 4260
2992
+ },
2993
+ {
2994
+ "epoch": 15.932835820895523,
2995
+ "grad_norm": 0.18404732644557953,
2996
+ "learning_rate": 9.286782436297073e-05,
2997
+ "loss": 0.0402,
2998
+ "step": 4270
2999
+ },
3000
+ {
3001
+ "epoch": 15.970149253731343,
3002
+ "grad_norm": 0.21722930669784546,
3003
+ "learning_rate": 9.282521182862629e-05,
3004
+ "loss": 0.0397,
3005
+ "step": 4280
3006
+ },
3007
+ {
3008
+ "epoch": 16.007462686567163,
3009
+ "grad_norm": 0.2523709833621979,
3010
+ "learning_rate": 9.278248221178798e-05,
3011
+ "loss": 0.0427,
3012
+ "step": 4290
3013
+ },
3014
+ {
3015
+ "epoch": 16.044776119402986,
3016
+ "grad_norm": 0.17736563086509705,
3017
+ "learning_rate": 9.273963562927695e-05,
3018
+ "loss": 0.0458,
3019
+ "step": 4300
3020
+ },
3021
+ {
3022
+ "epoch": 16.082089552238806,
3023
+ "grad_norm": 0.20613858103752136,
3024
+ "learning_rate": 9.269667219823412e-05,
3025
+ "loss": 0.0387,
3026
+ "step": 4310
3027
+ },
3028
+ {
3029
+ "epoch": 16.119402985074625,
3030
+ "grad_norm": 0.16557513177394867,
3031
+ "learning_rate": 9.265359203611987e-05,
3032
+ "loss": 0.0411,
3033
+ "step": 4320
3034
+ },
3035
+ {
3036
+ "epoch": 16.15671641791045,
3037
+ "grad_norm": 0.28119519352912903,
3038
+ "learning_rate": 9.261039526071374e-05,
3039
+ "loss": 0.0468,
3040
+ "step": 4330
3041
+ },
3042
+ {
3043
+ "epoch": 16.19402985074627,
3044
+ "grad_norm": 0.21538576483726501,
3045
+ "learning_rate": 9.256708199011401e-05,
3046
+ "loss": 0.0368,
3047
+ "step": 4340
3048
+ },
3049
+ {
3050
+ "epoch": 16.23134328358209,
3051
+ "grad_norm": 0.19657357037067413,
3052
+ "learning_rate": 9.252365234273755e-05,
3053
+ "loss": 0.038,
3054
+ "step": 4350
3055
+ },
3056
+ {
3057
+ "epoch": 16.26865671641791,
3058
+ "grad_norm": 0.19258421659469604,
3059
+ "learning_rate": 9.248010643731935e-05,
3060
+ "loss": 0.0414,
3061
+ "step": 4360
3062
+ },
3063
+ {
3064
+ "epoch": 16.30597014925373,
3065
+ "grad_norm": 0.28801625967025757,
3066
+ "learning_rate": 9.243644439291223e-05,
3067
+ "loss": 0.0387,
3068
+ "step": 4370
3069
+ },
3070
+ {
3071
+ "epoch": 16.34328358208955,
3072
+ "grad_norm": 0.16581468284130096,
3073
+ "learning_rate": 9.239266632888659e-05,
3074
+ "loss": 0.0383,
3075
+ "step": 4380
3076
+ },
3077
+ {
3078
+ "epoch": 16.380597014925375,
3079
+ "grad_norm": 0.34664949774742126,
3080
+ "learning_rate": 9.234877236492997e-05,
3081
+ "loss": 0.0453,
3082
+ "step": 4390
3083
+ },
3084
+ {
3085
+ "epoch": 16.417910447761194,
3086
+ "grad_norm": 0.1439947783946991,
3087
+ "learning_rate": 9.230476262104677e-05,
3088
+ "loss": 0.0466,
3089
+ "step": 4400
3090
+ },
3091
+ {
3092
+ "epoch": 16.455223880597014,
3093
+ "grad_norm": 0.15509940683841705,
3094
+ "learning_rate": 9.226063721755799e-05,
3095
+ "loss": 0.0488,
3096
+ "step": 4410
3097
+ },
3098
+ {
3099
+ "epoch": 16.492537313432837,
3100
+ "grad_norm": 0.18005985021591187,
3101
+ "learning_rate": 9.221639627510076e-05,
3102
+ "loss": 0.0407,
3103
+ "step": 4420
3104
+ },
3105
+ {
3106
+ "epoch": 16.529850746268657,
3107
+ "grad_norm": 0.16012470424175262,
3108
+ "learning_rate": 9.217203991462815e-05,
3109
+ "loss": 0.0394,
3110
+ "step": 4430
3111
+ },
3112
+ {
3113
+ "epoch": 16.567164179104477,
3114
+ "grad_norm": 0.2978847920894623,
3115
+ "learning_rate": 9.212756825740873e-05,
3116
+ "loss": 0.0451,
3117
+ "step": 4440
3118
+ },
3119
+ {
3120
+ "epoch": 16.604477611940297,
3121
+ "grad_norm": 0.2236834019422531,
3122
+ "learning_rate": 9.208298142502636e-05,
3123
+ "loss": 0.0487,
3124
+ "step": 4450
3125
+ },
3126
+ {
3127
+ "epoch": 16.64179104477612,
3128
+ "grad_norm": 0.2686060667037964,
3129
+ "learning_rate": 9.20382795393797e-05,
3130
+ "loss": 0.0403,
3131
+ "step": 4460
3132
+ },
3133
+ {
3134
+ "epoch": 16.67910447761194,
3135
+ "grad_norm": 0.33534038066864014,
3136
+ "learning_rate": 9.199346272268199e-05,
3137
+ "loss": 0.0385,
3138
+ "step": 4470
3139
+ },
3140
+ {
3141
+ "epoch": 16.71641791044776,
3142
+ "grad_norm": 0.19250528514385223,
3143
+ "learning_rate": 9.194853109746074e-05,
3144
+ "loss": 0.0441,
3145
+ "step": 4480
3146
+ },
3147
+ {
3148
+ "epoch": 16.753731343283583,
3149
+ "grad_norm": 0.19218407571315765,
3150
+ "learning_rate": 9.190348478655724e-05,
3151
+ "loss": 0.0474,
3152
+ "step": 4490
3153
+ },
3154
+ {
3155
+ "epoch": 16.791044776119403,
3156
+ "grad_norm": 0.21163488924503326,
3157
+ "learning_rate": 9.185832391312644e-05,
3158
+ "loss": 0.0411,
3159
+ "step": 4500
3160
+ },
3161
+ {
3162
+ "epoch": 16.828358208955223,
3163
+ "grad_norm": 0.1758819818496704,
3164
+ "learning_rate": 9.18130486006364e-05,
3165
+ "loss": 0.0462,
3166
+ "step": 4510
3167
+ },
3168
+ {
3169
+ "epoch": 16.865671641791046,
3170
+ "grad_norm": 0.18571069836616516,
3171
+ "learning_rate": 9.176765897286813e-05,
3172
+ "loss": 0.0425,
3173
+ "step": 4520
3174
+ },
3175
+ {
3176
+ "epoch": 16.902985074626866,
3177
+ "grad_norm": 0.20819155871868134,
3178
+ "learning_rate": 9.17221551539151e-05,
3179
+ "loss": 0.0428,
3180
+ "step": 4530
3181
+ },
3182
+ {
3183
+ "epoch": 16.940298507462686,
3184
+ "grad_norm": 0.30357328057289124,
3185
+ "learning_rate": 9.167653726818305e-05,
3186
+ "loss": 0.0414,
3187
+ "step": 4540
3188
+ },
3189
+ {
3190
+ "epoch": 16.97761194029851,
3191
+ "grad_norm": 0.20977462828159332,
3192
+ "learning_rate": 9.163080544038952e-05,
3193
+ "loss": 0.0447,
3194
+ "step": 4550
3195
+ },
3196
+ {
3197
+ "epoch": 17.01492537313433,
3198
+ "grad_norm": 0.2535971701145172,
3199
+ "learning_rate": 9.158495979556358e-05,
3200
+ "loss": 0.0384,
3201
+ "step": 4560
3202
+ },
3203
+ {
3204
+ "epoch": 17.05223880597015,
3205
+ "grad_norm": 0.2789897620677948,
3206
+ "learning_rate": 9.153900045904549e-05,
3207
+ "loss": 0.042,
3208
+ "step": 4570
3209
+ },
3210
+ {
3211
+ "epoch": 17.08955223880597,
3212
+ "grad_norm": 0.18474848568439484,
3213
+ "learning_rate": 9.14929275564863e-05,
3214
+ "loss": 0.0398,
3215
+ "step": 4580
3216
+ },
3217
+ {
3218
+ "epoch": 17.12686567164179,
3219
+ "grad_norm": 0.12615208327770233,
3220
+ "learning_rate": 9.144674121384757e-05,
3221
+ "loss": 0.0466,
3222
+ "step": 4590
3223
+ },
3224
+ {
3225
+ "epoch": 17.16417910447761,
3226
+ "grad_norm": 0.17756640911102295,
3227
+ "learning_rate": 9.140044155740101e-05,
3228
+ "loss": 0.035,
3229
+ "step": 4600
3230
+ },
3231
+ {
3232
+ "epoch": 17.20149253731343,
3233
+ "grad_norm": 0.24410821497440338,
3234
+ "learning_rate": 9.135402871372808e-05,
3235
+ "loss": 0.0459,
3236
+ "step": 4610
3237
+ },
3238
+ {
3239
+ "epoch": 17.238805970149254,
3240
+ "grad_norm": 0.21573011577129364,
3241
+ "learning_rate": 9.130750280971978e-05,
3242
+ "loss": 0.0385,
3243
+ "step": 4620
3244
+ },
3245
+ {
3246
+ "epoch": 17.276119402985074,
3247
+ "grad_norm": 0.13879653811454773,
3248
+ "learning_rate": 9.126086397257612e-05,
3249
+ "loss": 0.0391,
3250
+ "step": 4630
3251
+ },
3252
+ {
3253
+ "epoch": 17.313432835820894,
3254
+ "grad_norm": 0.17508305609226227,
3255
+ "learning_rate": 9.121411232980588e-05,
3256
+ "loss": 0.038,
3257
+ "step": 4640
3258
+ },
3259
+ {
3260
+ "epoch": 17.350746268656717,
3261
+ "grad_norm": 0.2536008358001709,
3262
+ "learning_rate": 9.116724800922629e-05,
3263
+ "loss": 0.0418,
3264
+ "step": 4650
3265
+ },
3266
+ {
3267
+ "epoch": 17.388059701492537,
3268
+ "grad_norm": 0.1942976713180542,
3269
+ "learning_rate": 9.112027113896262e-05,
3270
+ "loss": 0.052,
3271
+ "step": 4660
3272
+ },
3273
+ {
3274
+ "epoch": 17.425373134328357,
3275
+ "grad_norm": 0.16561119258403778,
3276
+ "learning_rate": 9.107318184744781e-05,
3277
+ "loss": 0.0451,
3278
+ "step": 4670
3279
+ },
3280
+ {
3281
+ "epoch": 17.46268656716418,
3282
+ "grad_norm": 0.22971832752227783,
3283
+ "learning_rate": 9.102598026342222e-05,
3284
+ "loss": 0.0407,
3285
+ "step": 4680
3286
+ },
3287
+ {
3288
+ "epoch": 17.5,
3289
+ "grad_norm": 0.1306753158569336,
3290
+ "learning_rate": 9.097866651593317e-05,
3291
+ "loss": 0.042,
3292
+ "step": 4690
3293
+ },
3294
+ {
3295
+ "epoch": 17.53731343283582,
3296
+ "grad_norm": 0.21278400719165802,
3297
+ "learning_rate": 9.093124073433463e-05,
3298
+ "loss": 0.0458,
3299
+ "step": 4700
3300
+ },
3301
+ {
3302
+ "epoch": 17.574626865671643,
3303
+ "grad_norm": 0.22757171094417572,
3304
+ "learning_rate": 9.088370304828685e-05,
3305
+ "loss": 0.0364,
3306
+ "step": 4710
3307
+ },
3308
+ {
3309
+ "epoch": 17.611940298507463,
3310
+ "grad_norm": 0.216596320271492,
3311
+ "learning_rate": 9.083605358775612e-05,
3312
+ "loss": 0.0434,
3313
+ "step": 4720
3314
+ },
3315
+ {
3316
+ "epoch": 17.649253731343283,
3317
+ "grad_norm": 0.13022471964359283,
3318
+ "learning_rate": 9.078829248301417e-05,
3319
+ "loss": 0.0415,
3320
+ "step": 4730
3321
+ },
3322
+ {
3323
+ "epoch": 17.686567164179106,
3324
+ "grad_norm": 0.2280716598033905,
3325
+ "learning_rate": 9.074041986463808e-05,
3326
+ "loss": 0.0385,
3327
+ "step": 4740
3328
+ },
3329
+ {
3330
+ "epoch": 17.723880597014926,
3331
+ "grad_norm": 0.14666135609149933,
3332
+ "learning_rate": 9.069243586350975e-05,
3333
+ "loss": 0.0347,
3334
+ "step": 4750
3335
+ },
3336
+ {
3337
+ "epoch": 17.761194029850746,
3338
+ "grad_norm": 0.1631281077861786,
3339
+ "learning_rate": 9.064434061081562e-05,
3340
+ "loss": 0.0407,
3341
+ "step": 4760
3342
+ },
3343
+ {
3344
+ "epoch": 17.798507462686565,
3345
+ "grad_norm": 0.18697327375411987,
3346
+ "learning_rate": 9.059613423804623e-05,
3347
+ "loss": 0.0425,
3348
+ "step": 4770
3349
+ },
3350
+ {
3351
+ "epoch": 17.83582089552239,
3352
+ "grad_norm": 0.12955111265182495,
3353
+ "learning_rate": 9.0547816876996e-05,
3354
+ "loss": 0.0417,
3355
+ "step": 4780
3356
+ },
3357
+ {
3358
+ "epoch": 17.87313432835821,
3359
+ "grad_norm": 0.15547148883342743,
3360
+ "learning_rate": 9.049938865976275e-05,
3361
+ "loss": 0.0409,
3362
+ "step": 4790
3363
+ },
3364
+ {
3365
+ "epoch": 17.91044776119403,
3366
+ "grad_norm": 0.1900598704814911,
3367
+ "learning_rate": 9.045084971874738e-05,
3368
+ "loss": 0.0369,
3369
+ "step": 4800
3370
+ },
3371
+ {
3372
+ "epoch": 17.94776119402985,
3373
+ "grad_norm": 0.1846715807914734,
3374
+ "learning_rate": 9.040220018665347e-05,
3375
+ "loss": 0.0415,
3376
+ "step": 4810
3377
+ },
3378
+ {
3379
+ "epoch": 17.98507462686567,
3380
+ "grad_norm": 0.1829937845468521,
3381
+ "learning_rate": 9.035344019648702e-05,
3382
+ "loss": 0.0407,
3383
+ "step": 4820
3384
+ },
3385
+ {
3386
+ "epoch": 18.02238805970149,
3387
+ "grad_norm": 0.25900354981422424,
3388
+ "learning_rate": 9.030456988155596e-05,
3389
+ "loss": 0.0398,
3390
+ "step": 4830
3391
+ },
3392
+ {
3393
+ "epoch": 18.059701492537314,
3394
+ "grad_norm": 0.21235992014408112,
3395
+ "learning_rate": 9.025558937546988e-05,
3396
+ "loss": 0.0477,
3397
+ "step": 4840
3398
+ },
3399
+ {
3400
+ "epoch": 18.097014925373134,
3401
+ "grad_norm": 0.18785078823566437,
3402
+ "learning_rate": 9.020649881213958e-05,
3403
+ "loss": 0.039,
3404
+ "step": 4850
3405
+ },
3406
+ {
3407
+ "epoch": 18.134328358208954,
3408
+ "grad_norm": 0.1951548010110855,
3409
+ "learning_rate": 9.015729832577681e-05,
3410
+ "loss": 0.0357,
3411
+ "step": 4860
3412
+ },
3413
+ {
3414
+ "epoch": 18.171641791044777,
3415
+ "grad_norm": 0.1280934363603592,
3416
+ "learning_rate": 9.010798805089384e-05,
3417
+ "loss": 0.0425,
3418
+ "step": 4870
3419
+ },
3420
+ {
3421
+ "epoch": 18.208955223880597,
3422
+ "grad_norm": 0.1693423092365265,
3423
+ "learning_rate": 9.005856812230304e-05,
3424
+ "loss": 0.0447,
3425
+ "step": 4880
3426
+ },
3427
+ {
3428
+ "epoch": 18.246268656716417,
3429
+ "grad_norm": 0.23712658882141113,
3430
+ "learning_rate": 9.000903867511666e-05,
3431
+ "loss": 0.042,
3432
+ "step": 4890
3433
+ },
3434
+ {
3435
+ "epoch": 18.28358208955224,
3436
+ "grad_norm": 0.26489710807800293,
3437
+ "learning_rate": 8.995939984474624e-05,
3438
+ "loss": 0.0457,
3439
+ "step": 4900
3440
+ },
3441
+ {
3442
+ "epoch": 18.32089552238806,
3443
+ "grad_norm": 0.20792756974697113,
3444
+ "learning_rate": 8.990965176690252e-05,
3445
+ "loss": 0.0422,
3446
+ "step": 4910
3447
+ },
3448
+ {
3449
+ "epoch": 18.35820895522388,
3450
+ "grad_norm": 0.18526089191436768,
3451
+ "learning_rate": 8.98597945775948e-05,
3452
+ "loss": 0.0366,
3453
+ "step": 4920
3454
+ },
3455
+ {
3456
+ "epoch": 18.395522388059703,
3457
+ "grad_norm": 0.2214607298374176,
3458
+ "learning_rate": 8.980982841313074e-05,
3459
+ "loss": 0.0405,
3460
+ "step": 4930
3461
+ },
3462
+ {
3463
+ "epoch": 18.432835820895523,
3464
+ "grad_norm": 0.1896953135728836,
3465
+ "learning_rate": 8.975975341011596e-05,
3466
+ "loss": 0.0391,
3467
+ "step": 4940
3468
+ },
3469
+ {
3470
+ "epoch": 18.470149253731343,
3471
+ "grad_norm": 0.1430232971906662,
3472
+ "learning_rate": 8.970956970545355e-05,
3473
+ "loss": 0.0403,
3474
+ "step": 4950
3475
+ },
3476
+ {
3477
+ "epoch": 18.507462686567163,
3478
+ "grad_norm": 0.1991272121667862,
3479
+ "learning_rate": 8.965927743634391e-05,
3480
+ "loss": 0.0429,
3481
+ "step": 4960
3482
+ },
3483
+ {
3484
+ "epoch": 18.544776119402986,
3485
+ "grad_norm": 0.2361849844455719,
3486
+ "learning_rate": 8.96088767402841e-05,
3487
+ "loss": 0.0416,
3488
+ "step": 4970
3489
+ },
3490
+ {
3491
+ "epoch": 18.582089552238806,
3492
+ "grad_norm": 0.25857019424438477,
3493
+ "learning_rate": 8.955836775506776e-05,
3494
+ "loss": 0.0461,
3495
+ "step": 4980
3496
+ },
3497
+ {
3498
+ "epoch": 18.619402985074625,
3499
+ "grad_norm": 0.12873682379722595,
3500
+ "learning_rate": 8.950775061878453e-05,
3501
+ "loss": 0.035,
3502
+ "step": 4990
3503
+ },
3504
+ {
3505
+ "epoch": 18.65671641791045,
3506
+ "grad_norm": 0.19786769151687622,
3507
+ "learning_rate": 8.945702546981969e-05,
3508
+ "loss": 0.0399,
3509
+ "step": 5000
3510
+ }
3511
+ ],
3512
+ "logging_steps": 10,
3513
+ "max_steps": 20000,
3514
+ "num_input_tokens_seen": 0,
3515
+ "num_train_epochs": 75,
3516
+ "save_steps": 500,
3517
+ "stateful_callbacks": {
3518
+ "TrainerControl": {
3519
+ "args": {
3520
+ "should_epoch_stop": false,
3521
+ "should_evaluate": false,
3522
+ "should_log": false,
3523
+ "should_save": true,
3524
+ "should_training_stop": false
3525
+ },
3526
+ "attributes": {}
3527
+ }
3528
+ },
3529
+ "total_flos": 3.758678370802253e+17,
3530
+ "train_batch_size": 32,
3531
+ "trial_name": null,
3532
+ "trial_params": null
3533
+ }
checkpoint-5500/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-5500/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-5500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af956da5fc247034f8acc05baa83b16179bfe24b3f0ae5b6ffa3a7aea4db48b5
3
+ size 123328576
checkpoint-5500/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-5500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:824f4483f101baf3361f5c6ba11df31164f64392f31713e19eaaebc052b7ca63
3
+ size 246824634
checkpoint-5500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69d4aa547bd6e670e9fd947ee26f6c4619dee2f8046c1b446067ae49dc3a9c0d
3
+ size 14244
checkpoint-5500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1c3fd1b661e3333c011abb51039c1da9b026667d8841fc5e277a18e72de1fd6
3
+ size 1064
checkpoint-5500/trainer_state.json ADDED
@@ -0,0 +1,3883 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 20.52238805970149,
5
+ "eval_steps": 500,
6
+ "global_step": 5500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03731343283582089,
13
+ "grad_norm": 0.8186072111129761,
14
+ "learning_rate": 1.0000000000000002e-06,
15
+ "loss": 1.3847,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.07462686567164178,
20
+ "grad_norm": 0.5007426142692566,
21
+ "learning_rate": 2.0000000000000003e-06,
22
+ "loss": 1.4283,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.11194029850746269,
27
+ "grad_norm": 0.49460887908935547,
28
+ "learning_rate": 3e-06,
29
+ "loss": 1.4868,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.14925373134328357,
34
+ "grad_norm": 0.5032920837402344,
35
+ "learning_rate": 4.000000000000001e-06,
36
+ "loss": 1.4491,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.1865671641791045,
41
+ "grad_norm": 0.5688469409942627,
42
+ "learning_rate": 5e-06,
43
+ "loss": 1.3703,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.22388059701492538,
48
+ "grad_norm": 0.5052517652511597,
49
+ "learning_rate": 6e-06,
50
+ "loss": 1.419,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.26119402985074625,
55
+ "grad_norm": 0.6315643787384033,
56
+ "learning_rate": 7.000000000000001e-06,
57
+ "loss": 1.3058,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.29850746268656714,
62
+ "grad_norm": 0.6060447692871094,
63
+ "learning_rate": 8.000000000000001e-06,
64
+ "loss": 1.2908,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.3358208955223881,
69
+ "grad_norm": 0.5513179302215576,
70
+ "learning_rate": 9e-06,
71
+ "loss": 1.2311,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.373134328358209,
76
+ "grad_norm": 0.8467404246330261,
77
+ "learning_rate": 1e-05,
78
+ "loss": 1.2043,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.41044776119402987,
83
+ "grad_norm": 0.8141824007034302,
84
+ "learning_rate": 1.1000000000000001e-05,
85
+ "loss": 1.0707,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.44776119402985076,
90
+ "grad_norm": 0.7932347059249878,
91
+ "learning_rate": 1.2e-05,
92
+ "loss": 0.9377,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.48507462686567165,
97
+ "grad_norm": 0.684220552444458,
98
+ "learning_rate": 1.3000000000000001e-05,
99
+ "loss": 0.714,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.5223880597014925,
104
+ "grad_norm": 0.5886895060539246,
105
+ "learning_rate": 1.4000000000000001e-05,
106
+ "loss": 0.6479,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5597014925373134,
111
+ "grad_norm": 0.4764939248561859,
112
+ "learning_rate": 1.5e-05,
113
+ "loss": 0.5463,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.5970149253731343,
118
+ "grad_norm": 0.4621008038520813,
119
+ "learning_rate": 1.6000000000000003e-05,
120
+ "loss": 0.4641,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.6343283582089553,
125
+ "grad_norm": 0.46492910385131836,
126
+ "learning_rate": 1.7000000000000003e-05,
127
+ "loss": 0.4159,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6716417910447762,
132
+ "grad_norm": 0.5017415881156921,
133
+ "learning_rate": 1.8e-05,
134
+ "loss": 0.4094,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.7089552238805971,
139
+ "grad_norm": 0.34392210841178894,
140
+ "learning_rate": 1.9e-05,
141
+ "loss": 0.3478,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.746268656716418,
146
+ "grad_norm": 0.3240516483783722,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3821,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7835820895522388,
153
+ "grad_norm": 0.26301339268684387,
154
+ "learning_rate": 2.1e-05,
155
+ "loss": 0.3606,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.8208955223880597,
160
+ "grad_norm": 0.34712520241737366,
161
+ "learning_rate": 2.2000000000000003e-05,
162
+ "loss": 0.3421,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.8582089552238806,
167
+ "grad_norm": 0.3248469829559326,
168
+ "learning_rate": 2.3000000000000003e-05,
169
+ "loss": 0.3389,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.8955223880597015,
174
+ "grad_norm": 0.298149436712265,
175
+ "learning_rate": 2.4e-05,
176
+ "loss": 0.3145,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.9328358208955224,
181
+ "grad_norm": 0.2757190763950348,
182
+ "learning_rate": 2.5e-05,
183
+ "loss": 0.3065,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.9701492537313433,
188
+ "grad_norm": 0.30510950088500977,
189
+ "learning_rate": 2.6000000000000002e-05,
190
+ "loss": 0.2971,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.007462686567164,
195
+ "grad_norm": 0.37349891662597656,
196
+ "learning_rate": 2.7000000000000002e-05,
197
+ "loss": 0.3273,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.044776119402985,
202
+ "grad_norm": 0.3667634129524231,
203
+ "learning_rate": 2.8000000000000003e-05,
204
+ "loss": 0.308,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.0820895522388059,
209
+ "grad_norm": 0.3463355004787445,
210
+ "learning_rate": 2.9e-05,
211
+ "loss": 0.3109,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.1194029850746268,
216
+ "grad_norm": 0.3888525366783142,
217
+ "learning_rate": 3e-05,
218
+ "loss": 0.2644,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.1567164179104479,
223
+ "grad_norm": 0.3749147951602936,
224
+ "learning_rate": 3.1e-05,
225
+ "loss": 0.2858,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.1940298507462686,
230
+ "grad_norm": 0.3270276188850403,
231
+ "learning_rate": 3.2000000000000005e-05,
232
+ "loss": 0.2573,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.2313432835820897,
237
+ "grad_norm": 0.3658592998981476,
238
+ "learning_rate": 3.3e-05,
239
+ "loss": 0.2613,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.2686567164179103,
244
+ "grad_norm": 0.3526328206062317,
245
+ "learning_rate": 3.4000000000000007e-05,
246
+ "loss": 0.2328,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.3059701492537314,
251
+ "grad_norm": 0.4528139531612396,
252
+ "learning_rate": 3.5e-05,
253
+ "loss": 0.2429,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.3432835820895521,
258
+ "grad_norm": 0.5426791310310364,
259
+ "learning_rate": 3.6e-05,
260
+ "loss": 0.2209,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.3805970149253732,
265
+ "grad_norm": 0.41844552755355835,
266
+ "learning_rate": 3.7e-05,
267
+ "loss": 0.2319,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.417910447761194,
272
+ "grad_norm": 0.4749431908130646,
273
+ "learning_rate": 3.8e-05,
274
+ "loss": 0.2233,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.455223880597015,
279
+ "grad_norm": 0.7010189890861511,
280
+ "learning_rate": 3.9000000000000006e-05,
281
+ "loss": 0.2181,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.4925373134328357,
286
+ "grad_norm": 0.5747635960578918,
287
+ "learning_rate": 4e-05,
288
+ "loss": 0.213,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.5298507462686568,
293
+ "grad_norm": 0.3661474287509918,
294
+ "learning_rate": 4.1e-05,
295
+ "loss": 0.2171,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.5671641791044775,
300
+ "grad_norm": 0.467835396528244,
301
+ "learning_rate": 4.2e-05,
302
+ "loss": 0.1985,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.6044776119402986,
307
+ "grad_norm": 0.5470123291015625,
308
+ "learning_rate": 4.3e-05,
309
+ "loss": 0.2176,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.6417910447761193,
314
+ "grad_norm": 0.5761199593544006,
315
+ "learning_rate": 4.4000000000000006e-05,
316
+ "loss": 0.2007,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.6791044776119404,
321
+ "grad_norm": 0.48257485032081604,
322
+ "learning_rate": 4.5e-05,
323
+ "loss": 0.2043,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.716417910447761,
328
+ "grad_norm": 0.48353052139282227,
329
+ "learning_rate": 4.600000000000001e-05,
330
+ "loss": 0.1872,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.7537313432835822,
335
+ "grad_norm": 0.4388391375541687,
336
+ "learning_rate": 4.7e-05,
337
+ "loss": 0.206,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.7910447761194028,
342
+ "grad_norm": 0.47332626581192017,
343
+ "learning_rate": 4.8e-05,
344
+ "loss": 0.1876,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.828358208955224,
349
+ "grad_norm": 0.8053535223007202,
350
+ "learning_rate": 4.9e-05,
351
+ "loss": 0.1839,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 1.8656716417910446,
356
+ "grad_norm": 0.413979709148407,
357
+ "learning_rate": 5e-05,
358
+ "loss": 0.1732,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 1.9029850746268657,
363
+ "grad_norm": 0.36910712718963623,
364
+ "learning_rate": 5.1000000000000006e-05,
365
+ "loss": 0.1827,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 1.9402985074626866,
370
+ "grad_norm": 0.8458298444747925,
371
+ "learning_rate": 5.2000000000000004e-05,
372
+ "loss": 0.1727,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 1.9776119402985075,
377
+ "grad_norm": 0.5452115535736084,
378
+ "learning_rate": 5.300000000000001e-05,
379
+ "loss": 0.1818,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 2.014925373134328,
384
+ "grad_norm": 0.4518108069896698,
385
+ "learning_rate": 5.4000000000000005e-05,
386
+ "loss": 0.177,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 2.0522388059701493,
391
+ "grad_norm": 0.66865074634552,
392
+ "learning_rate": 5.500000000000001e-05,
393
+ "loss": 0.1726,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 2.08955223880597,
398
+ "grad_norm": 0.6536034345626831,
399
+ "learning_rate": 5.6000000000000006e-05,
400
+ "loss": 0.1541,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 2.126865671641791,
405
+ "grad_norm": 0.5571377277374268,
406
+ "learning_rate": 5.6999999999999996e-05,
407
+ "loss": 0.1671,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 2.1641791044776117,
412
+ "grad_norm": 0.5385546684265137,
413
+ "learning_rate": 5.8e-05,
414
+ "loss": 0.1582,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 2.201492537313433,
419
+ "grad_norm": 0.577961266040802,
420
+ "learning_rate": 5.9e-05,
421
+ "loss": 0.1528,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 2.2388059701492535,
426
+ "grad_norm": 0.5082416534423828,
427
+ "learning_rate": 6e-05,
428
+ "loss": 0.1638,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 2.2761194029850746,
433
+ "grad_norm": 0.5490861535072327,
434
+ "learning_rate": 6.1e-05,
435
+ "loss": 0.166,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 2.3134328358208958,
440
+ "grad_norm": 0.492366760969162,
441
+ "learning_rate": 6.2e-05,
442
+ "loss": 0.1481,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 2.3507462686567164,
447
+ "grad_norm": 0.3702855110168457,
448
+ "learning_rate": 6.3e-05,
449
+ "loss": 0.1514,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 2.388059701492537,
454
+ "grad_norm": 0.664667010307312,
455
+ "learning_rate": 6.400000000000001e-05,
456
+ "loss": 0.1441,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 2.425373134328358,
461
+ "grad_norm": 0.33382174372673035,
462
+ "learning_rate": 6.500000000000001e-05,
463
+ "loss": 0.1573,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 2.4626865671641793,
468
+ "grad_norm": 0.4848814010620117,
469
+ "learning_rate": 6.6e-05,
470
+ "loss": 0.1457,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 2.5,
475
+ "grad_norm": 0.3649997413158417,
476
+ "learning_rate": 6.7e-05,
477
+ "loss": 0.1467,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 2.5373134328358207,
482
+ "grad_norm": 0.6385223865509033,
483
+ "learning_rate": 6.800000000000001e-05,
484
+ "loss": 0.145,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 2.574626865671642,
489
+ "grad_norm": 0.4580625891685486,
490
+ "learning_rate": 6.9e-05,
491
+ "loss": 0.1352,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 2.611940298507463,
496
+ "grad_norm": 0.5141746401786804,
497
+ "learning_rate": 7e-05,
498
+ "loss": 0.1444,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 2.6492537313432836,
503
+ "grad_norm": 0.40220722556114197,
504
+ "learning_rate": 7.1e-05,
505
+ "loss": 0.1493,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 2.6865671641791042,
510
+ "grad_norm": 0.5510571002960205,
511
+ "learning_rate": 7.2e-05,
512
+ "loss": 0.1387,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 2.7238805970149254,
517
+ "grad_norm": 0.43814659118652344,
518
+ "learning_rate": 7.3e-05,
519
+ "loss": 0.1374,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 2.7611940298507465,
524
+ "grad_norm": 0.4118008613586426,
525
+ "learning_rate": 7.4e-05,
526
+ "loss": 0.1297,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 2.798507462686567,
531
+ "grad_norm": 0.5626503229141235,
532
+ "learning_rate": 7.500000000000001e-05,
533
+ "loss": 0.1299,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 2.835820895522388,
538
+ "grad_norm": 0.4066360592842102,
539
+ "learning_rate": 7.6e-05,
540
+ "loss": 0.1102,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 2.873134328358209,
545
+ "grad_norm": 0.47184985876083374,
546
+ "learning_rate": 7.7e-05,
547
+ "loss": 0.1219,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 2.91044776119403,
552
+ "grad_norm": 0.6611475348472595,
553
+ "learning_rate": 7.800000000000001e-05,
554
+ "loss": 0.1267,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 2.9477611940298507,
559
+ "grad_norm": 0.3570108413696289,
560
+ "learning_rate": 7.900000000000001e-05,
561
+ "loss": 0.1191,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 2.9850746268656714,
566
+ "grad_norm": 0.4581681489944458,
567
+ "learning_rate": 8e-05,
568
+ "loss": 0.1209,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 3.0223880597014925,
573
+ "grad_norm": 0.4643435776233673,
574
+ "learning_rate": 8.1e-05,
575
+ "loss": 0.129,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 3.0597014925373136,
580
+ "grad_norm": 0.5595763921737671,
581
+ "learning_rate": 8.2e-05,
582
+ "loss": 0.1158,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 3.0970149253731343,
587
+ "grad_norm": 0.48848605155944824,
588
+ "learning_rate": 8.3e-05,
589
+ "loss": 0.1188,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 3.1343283582089554,
594
+ "grad_norm": 0.4496570825576782,
595
+ "learning_rate": 8.4e-05,
596
+ "loss": 0.114,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 3.171641791044776,
601
+ "grad_norm": 0.31364986300468445,
602
+ "learning_rate": 8.5e-05,
603
+ "loss": 0.1196,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 3.208955223880597,
608
+ "grad_norm": 0.3395878076553345,
609
+ "learning_rate": 8.6e-05,
610
+ "loss": 0.1124,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 3.246268656716418,
615
+ "grad_norm": 0.4917413592338562,
616
+ "learning_rate": 8.7e-05,
617
+ "loss": 0.1074,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 3.283582089552239,
622
+ "grad_norm": 0.44114553928375244,
623
+ "learning_rate": 8.800000000000001e-05,
624
+ "loss": 0.1095,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 3.3208955223880596,
629
+ "grad_norm": 0.3323831558227539,
630
+ "learning_rate": 8.900000000000001e-05,
631
+ "loss": 0.106,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 3.3582089552238807,
636
+ "grad_norm": 0.4495660066604614,
637
+ "learning_rate": 9e-05,
638
+ "loss": 0.1222,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 3.3955223880597014,
643
+ "grad_norm": 0.40784788131713867,
644
+ "learning_rate": 9.1e-05,
645
+ "loss": 0.1048,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 3.4328358208955225,
650
+ "grad_norm": 0.4643700420856476,
651
+ "learning_rate": 9.200000000000001e-05,
652
+ "loss": 0.1097,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 3.470149253731343,
657
+ "grad_norm": 0.472494512796402,
658
+ "learning_rate": 9.300000000000001e-05,
659
+ "loss": 0.1041,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 3.5074626865671643,
664
+ "grad_norm": 0.6110897660255432,
665
+ "learning_rate": 9.4e-05,
666
+ "loss": 0.0959,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 3.544776119402985,
671
+ "grad_norm": 0.5313069820404053,
672
+ "learning_rate": 9.5e-05,
673
+ "loss": 0.113,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 3.582089552238806,
678
+ "grad_norm": 0.4223133623600006,
679
+ "learning_rate": 9.6e-05,
680
+ "loss": 0.099,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 3.6194029850746268,
685
+ "grad_norm": 0.5464731454849243,
686
+ "learning_rate": 9.7e-05,
687
+ "loss": 0.1008,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 3.656716417910448,
692
+ "grad_norm": 0.3538314402103424,
693
+ "learning_rate": 9.8e-05,
694
+ "loss": 0.1049,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 3.6940298507462686,
699
+ "grad_norm": 0.7460148334503174,
700
+ "learning_rate": 9.900000000000001e-05,
701
+ "loss": 0.1088,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 3.7313432835820897,
706
+ "grad_norm": 0.3210597038269043,
707
+ "learning_rate": 0.0001,
708
+ "loss": 0.1041,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 3.7686567164179103,
713
+ "grad_norm": 0.4450497627258301,
714
+ "learning_rate": 9.999993165095463e-05,
715
+ "loss": 0.0985,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 3.8059701492537314,
720
+ "grad_norm": 0.4348960816860199,
721
+ "learning_rate": 9.999972660400536e-05,
722
+ "loss": 0.1015,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 3.843283582089552,
727
+ "grad_norm": 0.462782621383667,
728
+ "learning_rate": 9.999938485971279e-05,
729
+ "loss": 0.1068,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 3.8805970149253732,
734
+ "grad_norm": 0.3801368474960327,
735
+ "learning_rate": 9.999890641901125e-05,
736
+ "loss": 0.1117,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 3.917910447761194,
741
+ "grad_norm": 0.45135366916656494,
742
+ "learning_rate": 9.999829128320874e-05,
743
+ "loss": 0.0917,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 3.955223880597015,
748
+ "grad_norm": 0.41138389706611633,
749
+ "learning_rate": 9.999753945398704e-05,
750
+ "loss": 0.1049,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 3.9925373134328357,
755
+ "grad_norm": 0.4976252317428589,
756
+ "learning_rate": 9.999665093340165e-05,
757
+ "loss": 0.1029,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 4.029850746268656,
762
+ "grad_norm": 0.46372008323669434,
763
+ "learning_rate": 9.99956257238817e-05,
764
+ "loss": 0.1012,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 4.067164179104478,
769
+ "grad_norm": 0.546938955783844,
770
+ "learning_rate": 9.999446382823013e-05,
771
+ "loss": 0.0829,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 4.104477611940299,
776
+ "grad_norm": 0.40513405203819275,
777
+ "learning_rate": 9.999316524962345e-05,
778
+ "loss": 0.0933,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 4.141791044776119,
783
+ "grad_norm": 0.4198484420776367,
784
+ "learning_rate": 9.999172999161198e-05,
785
+ "loss": 0.0895,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 4.17910447761194,
790
+ "grad_norm": 0.3965628743171692,
791
+ "learning_rate": 9.999015805811965e-05,
792
+ "loss": 0.0917,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 4.2164179104477615,
797
+ "grad_norm": 0.3095884621143341,
798
+ "learning_rate": 9.998844945344405e-05,
799
+ "loss": 0.0953,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 4.253731343283582,
804
+ "grad_norm": 0.7962276339530945,
805
+ "learning_rate": 9.998660418225645e-05,
806
+ "loss": 0.0979,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 4.291044776119403,
811
+ "grad_norm": 0.42066490650177,
812
+ "learning_rate": 9.998462224960175e-05,
813
+ "loss": 0.099,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 4.3283582089552235,
818
+ "grad_norm": 0.3894193470478058,
819
+ "learning_rate": 9.998250366089848e-05,
820
+ "loss": 0.0887,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 4.365671641791045,
825
+ "grad_norm": 0.28998032212257385,
826
+ "learning_rate": 9.998024842193876e-05,
827
+ "loss": 0.0943,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 4.402985074626866,
832
+ "grad_norm": 0.3919823467731476,
833
+ "learning_rate": 9.997785653888835e-05,
834
+ "loss": 0.0916,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 4.440298507462686,
839
+ "grad_norm": 0.3708650469779968,
840
+ "learning_rate": 9.997532801828658e-05,
841
+ "loss": 0.0858,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 4.477611940298507,
846
+ "grad_norm": 0.2935069799423218,
847
+ "learning_rate": 9.997266286704631e-05,
848
+ "loss": 0.0992,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 4.514925373134329,
853
+ "grad_norm": 0.4675377607345581,
854
+ "learning_rate": 9.996986109245395e-05,
855
+ "loss": 0.0854,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 4.552238805970149,
860
+ "grad_norm": 0.31374865770339966,
861
+ "learning_rate": 9.996692270216947e-05,
862
+ "loss": 0.0788,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 4.58955223880597,
867
+ "grad_norm": 0.419249951839447,
868
+ "learning_rate": 9.996384770422629e-05,
869
+ "loss": 0.0873,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 4.6268656716417915,
874
+ "grad_norm": 0.26002731919288635,
875
+ "learning_rate": 9.996063610703137e-05,
876
+ "loss": 0.0845,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 4.664179104477612,
881
+ "grad_norm": 0.29573896527290344,
882
+ "learning_rate": 9.995728791936504e-05,
883
+ "loss": 0.091,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 4.701492537313433,
888
+ "grad_norm": 0.33090147376060486,
889
+ "learning_rate": 9.995380315038119e-05,
890
+ "loss": 0.0827,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 4.7388059701492535,
895
+ "grad_norm": 0.24417485296726227,
896
+ "learning_rate": 9.9950181809607e-05,
897
+ "loss": 0.0859,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 4.776119402985074,
902
+ "grad_norm": 0.48290401697158813,
903
+ "learning_rate": 9.994642390694308e-05,
904
+ "loss": 0.0889,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 4.813432835820896,
909
+ "grad_norm": 0.4479697048664093,
910
+ "learning_rate": 9.99425294526634e-05,
911
+ "loss": 0.097,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 4.850746268656716,
916
+ "grad_norm": 0.3560147285461426,
917
+ "learning_rate": 9.993849845741524e-05,
918
+ "loss": 0.0904,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 4.888059701492537,
923
+ "grad_norm": 0.6645416617393494,
924
+ "learning_rate": 9.99343309322192e-05,
925
+ "loss": 0.0922,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 4.925373134328359,
930
+ "grad_norm": 0.29696759581565857,
931
+ "learning_rate": 9.993002688846913e-05,
932
+ "loss": 0.093,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 4.962686567164179,
937
+ "grad_norm": 0.47146692872047424,
938
+ "learning_rate": 9.992558633793212e-05,
939
+ "loss": 0.085,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 5.0,
944
+ "grad_norm": 0.3430916368961334,
945
+ "learning_rate": 9.992100929274846e-05,
946
+ "loss": 0.0805,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 5.037313432835821,
951
+ "grad_norm": 0.3205055892467499,
952
+ "learning_rate": 9.991629576543163e-05,
953
+ "loss": 0.0766,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 5.074626865671641,
958
+ "grad_norm": 0.3664805293083191,
959
+ "learning_rate": 9.991144576886823e-05,
960
+ "loss": 0.0766,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 5.111940298507463,
965
+ "grad_norm": 0.3753412663936615,
966
+ "learning_rate": 9.990645931631796e-05,
967
+ "loss": 0.0688,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 5.149253731343284,
972
+ "grad_norm": 0.31633055210113525,
973
+ "learning_rate": 9.990133642141359e-05,
974
+ "loss": 0.0796,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 5.186567164179104,
979
+ "grad_norm": 0.3355732262134552,
980
+ "learning_rate": 9.989607709816091e-05,
981
+ "loss": 0.0716,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 5.223880597014926,
986
+ "grad_norm": 0.24850831925868988,
987
+ "learning_rate": 9.989068136093873e-05,
988
+ "loss": 0.0778,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 5.2611940298507465,
993
+ "grad_norm": 0.29537102580070496,
994
+ "learning_rate": 9.988514922449879e-05,
995
+ "loss": 0.0759,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 5.298507462686567,
1000
+ "grad_norm": 0.3430945873260498,
1001
+ "learning_rate": 9.987948070396571e-05,
1002
+ "loss": 0.0774,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 5.335820895522388,
1007
+ "grad_norm": 0.5220637917518616,
1008
+ "learning_rate": 9.987367581483705e-05,
1009
+ "loss": 0.0836,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 5.373134328358209,
1014
+ "grad_norm": 0.28184008598327637,
1015
+ "learning_rate": 9.986773457298311e-05,
1016
+ "loss": 0.0752,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 5.41044776119403,
1021
+ "grad_norm": 0.36261311173439026,
1022
+ "learning_rate": 9.986165699464705e-05,
1023
+ "loss": 0.075,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 5.447761194029851,
1028
+ "grad_norm": 0.5107380151748657,
1029
+ "learning_rate": 9.985544309644475e-05,
1030
+ "loss": 0.0814,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 5.485074626865671,
1035
+ "grad_norm": 0.2446671426296234,
1036
+ "learning_rate": 9.984909289536473e-05,
1037
+ "loss": 0.0704,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 5.522388059701493,
1042
+ "grad_norm": 0.30449381470680237,
1043
+ "learning_rate": 9.984260640876821e-05,
1044
+ "loss": 0.0794,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 5.559701492537314,
1049
+ "grad_norm": 0.25645050406455994,
1050
+ "learning_rate": 9.983598365438902e-05,
1051
+ "loss": 0.0709,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 5.597014925373134,
1056
+ "grad_norm": 0.23825006186962128,
1057
+ "learning_rate": 9.98292246503335e-05,
1058
+ "loss": 0.0828,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 5.634328358208955,
1063
+ "grad_norm": 0.3259269893169403,
1064
+ "learning_rate": 9.98223294150805e-05,
1065
+ "loss": 0.0824,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 5.6716417910447765,
1070
+ "grad_norm": 0.24058914184570312,
1071
+ "learning_rate": 9.981529796748134e-05,
1072
+ "loss": 0.073,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 5.708955223880597,
1077
+ "grad_norm": 0.34457242488861084,
1078
+ "learning_rate": 9.980813032675974e-05,
1079
+ "loss": 0.0845,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 5.746268656716418,
1084
+ "grad_norm": 0.32940393686294556,
1085
+ "learning_rate": 9.980082651251175e-05,
1086
+ "loss": 0.0832,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 5.7835820895522385,
1091
+ "grad_norm": 0.5683007836341858,
1092
+ "learning_rate": 9.979338654470569e-05,
1093
+ "loss": 0.0836,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 5.82089552238806,
1098
+ "grad_norm": 0.31041061878204346,
1099
+ "learning_rate": 9.97858104436822e-05,
1100
+ "loss": 0.07,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 5.858208955223881,
1105
+ "grad_norm": 0.37858131527900696,
1106
+ "learning_rate": 9.977809823015401e-05,
1107
+ "loss": 0.0738,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 5.895522388059701,
1112
+ "grad_norm": 0.2743091583251953,
1113
+ "learning_rate": 9.977024992520602e-05,
1114
+ "loss": 0.0761,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 5.932835820895522,
1119
+ "grad_norm": 0.29117098450660706,
1120
+ "learning_rate": 9.976226555029522e-05,
1121
+ "loss": 0.0777,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 5.970149253731344,
1126
+ "grad_norm": 0.31398633122444153,
1127
+ "learning_rate": 9.975414512725057e-05,
1128
+ "loss": 0.0664,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 6.007462686567164,
1133
+ "grad_norm": 0.2684272527694702,
1134
+ "learning_rate": 9.974588867827301e-05,
1135
+ "loss": 0.0686,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 6.044776119402985,
1140
+ "grad_norm": 0.3945397436618805,
1141
+ "learning_rate": 9.973749622593534e-05,
1142
+ "loss": 0.0614,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 6.082089552238806,
1147
+ "grad_norm": 0.2747954726219177,
1148
+ "learning_rate": 9.972896779318219e-05,
1149
+ "loss": 0.0681,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 6.119402985074627,
1154
+ "grad_norm": 0.43257200717926025,
1155
+ "learning_rate": 9.972030340333001e-05,
1156
+ "loss": 0.0725,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 6.156716417910448,
1161
+ "grad_norm": 0.3559250831604004,
1162
+ "learning_rate": 9.97115030800669e-05,
1163
+ "loss": 0.0804,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 6.1940298507462686,
1168
+ "grad_norm": 0.3079264760017395,
1169
+ "learning_rate": 9.970256684745258e-05,
1170
+ "loss": 0.0649,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 6.231343283582089,
1175
+ "grad_norm": 0.32298946380615234,
1176
+ "learning_rate": 9.969349472991838e-05,
1177
+ "loss": 0.0668,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 6.268656716417911,
1182
+ "grad_norm": 0.2826225459575653,
1183
+ "learning_rate": 9.968428675226714e-05,
1184
+ "loss": 0.0734,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 6.3059701492537314,
1189
+ "grad_norm": 0.39002349972724915,
1190
+ "learning_rate": 9.967494293967312e-05,
1191
+ "loss": 0.0728,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 6.343283582089552,
1196
+ "grad_norm": 0.403890997171402,
1197
+ "learning_rate": 9.966546331768191e-05,
1198
+ "loss": 0.067,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 6.380597014925373,
1203
+ "grad_norm": 0.3755359351634979,
1204
+ "learning_rate": 9.965584791221048e-05,
1205
+ "loss": 0.0755,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 6.417910447761194,
1210
+ "grad_norm": 0.26346635818481445,
1211
+ "learning_rate": 9.964609674954696e-05,
1212
+ "loss": 0.0728,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 6.455223880597015,
1217
+ "grad_norm": 0.45292145013809204,
1218
+ "learning_rate": 9.963620985635065e-05,
1219
+ "loss": 0.0731,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 6.492537313432836,
1224
+ "grad_norm": 0.3568434715270996,
1225
+ "learning_rate": 9.962618725965196e-05,
1226
+ "loss": 0.0761,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 6.529850746268656,
1231
+ "grad_norm": 0.2551257014274597,
1232
+ "learning_rate": 9.961602898685226e-05,
1233
+ "loss": 0.0694,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 6.567164179104478,
1238
+ "grad_norm": 0.6106354594230652,
1239
+ "learning_rate": 9.96057350657239e-05,
1240
+ "loss": 0.0827,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 6.604477611940299,
1245
+ "grad_norm": 0.3226093053817749,
1246
+ "learning_rate": 9.959530552441005e-05,
1247
+ "loss": 0.0716,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 6.641791044776119,
1252
+ "grad_norm": 0.4297254979610443,
1253
+ "learning_rate": 9.95847403914247e-05,
1254
+ "loss": 0.0748,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 6.67910447761194,
1259
+ "grad_norm": 0.26469680666923523,
1260
+ "learning_rate": 9.95740396956525e-05,
1261
+ "loss": 0.074,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 6.7164179104477615,
1266
+ "grad_norm": 0.22717897593975067,
1267
+ "learning_rate": 9.956320346634876e-05,
1268
+ "loss": 0.0739,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 6.753731343283582,
1273
+ "grad_norm": 0.4513498544692993,
1274
+ "learning_rate": 9.955223173313931e-05,
1275
+ "loss": 0.0664,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 6.791044776119403,
1280
+ "grad_norm": 0.31683439016342163,
1281
+ "learning_rate": 9.954112452602045e-05,
1282
+ "loss": 0.069,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 6.8283582089552235,
1287
+ "grad_norm": 0.3350532650947571,
1288
+ "learning_rate": 9.952988187535886e-05,
1289
+ "loss": 0.0699,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 6.865671641791045,
1294
+ "grad_norm": 0.29829463362693787,
1295
+ "learning_rate": 9.95185038118915e-05,
1296
+ "loss": 0.0663,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 6.902985074626866,
1301
+ "grad_norm": 0.31650781631469727,
1302
+ "learning_rate": 9.950699036672559e-05,
1303
+ "loss": 0.0668,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 6.940298507462686,
1308
+ "grad_norm": 0.360944926738739,
1309
+ "learning_rate": 9.949534157133844e-05,
1310
+ "loss": 0.0696,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 6.977611940298507,
1315
+ "grad_norm": 0.31337013840675354,
1316
+ "learning_rate": 9.948355745757741e-05,
1317
+ "loss": 0.073,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 7.014925373134329,
1322
+ "grad_norm": 0.4675919711589813,
1323
+ "learning_rate": 9.94716380576598e-05,
1324
+ "loss": 0.0688,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 7.052238805970149,
1329
+ "grad_norm": 0.3031919002532959,
1330
+ "learning_rate": 9.945958340417283e-05,
1331
+ "loss": 0.0596,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 7.08955223880597,
1336
+ "grad_norm": 0.24858474731445312,
1337
+ "learning_rate": 9.944739353007344e-05,
1338
+ "loss": 0.0717,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 7.126865671641791,
1343
+ "grad_norm": 0.20959483087062836,
1344
+ "learning_rate": 9.943506846868826e-05,
1345
+ "loss": 0.0694,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 7.164179104477612,
1350
+ "grad_norm": 0.35621434450149536,
1351
+ "learning_rate": 9.942260825371358e-05,
1352
+ "loss": 0.063,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 7.201492537313433,
1357
+ "grad_norm": 0.3462587594985962,
1358
+ "learning_rate": 9.941001291921512e-05,
1359
+ "loss": 0.068,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 7.2388059701492535,
1364
+ "grad_norm": 0.38649681210517883,
1365
+ "learning_rate": 9.939728249962807e-05,
1366
+ "loss": 0.0638,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 7.276119402985074,
1371
+ "grad_norm": 0.29564595222473145,
1372
+ "learning_rate": 9.938441702975689e-05,
1373
+ "loss": 0.0626,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 7.313432835820896,
1378
+ "grad_norm": 0.339857816696167,
1379
+ "learning_rate": 9.937141654477528e-05,
1380
+ "loss": 0.0535,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 7.350746268656716,
1385
+ "grad_norm": 0.2591215670108795,
1386
+ "learning_rate": 9.93582810802261e-05,
1387
+ "loss": 0.0645,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 7.388059701492537,
1392
+ "grad_norm": 0.30237796902656555,
1393
+ "learning_rate": 9.934501067202117e-05,
1394
+ "loss": 0.0675,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 7.425373134328359,
1399
+ "grad_norm": 0.28394174575805664,
1400
+ "learning_rate": 9.93316053564413e-05,
1401
+ "loss": 0.0643,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 7.462686567164179,
1406
+ "grad_norm": 0.3124663233757019,
1407
+ "learning_rate": 9.931806517013612e-05,
1408
+ "loss": 0.059,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 7.5,
1413
+ "grad_norm": 0.36073037981987,
1414
+ "learning_rate": 9.930439015012396e-05,
1415
+ "loss": 0.0606,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 7.537313432835821,
1420
+ "grad_norm": 0.4091481864452362,
1421
+ "learning_rate": 9.929058033379181e-05,
1422
+ "loss": 0.0603,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 7.574626865671641,
1427
+ "grad_norm": 0.44718074798583984,
1428
+ "learning_rate": 9.927663575889521e-05,
1429
+ "loss": 0.0741,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 7.611940298507463,
1434
+ "grad_norm": 0.3819601833820343,
1435
+ "learning_rate": 9.926255646355804e-05,
1436
+ "loss": 0.0707,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 7.649253731343284,
1441
+ "grad_norm": 0.23336420953273773,
1442
+ "learning_rate": 9.92483424862726e-05,
1443
+ "loss": 0.0676,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 7.686567164179104,
1448
+ "grad_norm": 0.24415315687656403,
1449
+ "learning_rate": 9.923399386589933e-05,
1450
+ "loss": 0.0594,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 7.723880597014926,
1455
+ "grad_norm": 0.3735473155975342,
1456
+ "learning_rate": 9.921951064166684e-05,
1457
+ "loss": 0.062,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 7.7611940298507465,
1462
+ "grad_norm": 0.31629472970962524,
1463
+ "learning_rate": 9.92048928531717e-05,
1464
+ "loss": 0.0606,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 7.798507462686567,
1469
+ "grad_norm": 0.37902557849884033,
1470
+ "learning_rate": 9.919014054037836e-05,
1471
+ "loss": 0.0584,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 7.835820895522388,
1476
+ "grad_norm": 0.3486720323562622,
1477
+ "learning_rate": 9.917525374361912e-05,
1478
+ "loss": 0.056,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 7.8731343283582085,
1483
+ "grad_norm": 0.3731362521648407,
1484
+ "learning_rate": 9.91602325035939e-05,
1485
+ "loss": 0.0601,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 7.91044776119403,
1490
+ "grad_norm": 0.3560399115085602,
1491
+ "learning_rate": 9.914507686137019e-05,
1492
+ "loss": 0.06,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 7.947761194029851,
1497
+ "grad_norm": 0.30075564980506897,
1498
+ "learning_rate": 9.912978685838294e-05,
1499
+ "loss": 0.0657,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 7.985074626865671,
1504
+ "grad_norm": 0.2984028458595276,
1505
+ "learning_rate": 9.911436253643445e-05,
1506
+ "loss": 0.0587,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 8.022388059701493,
1511
+ "grad_norm": 0.1980169117450714,
1512
+ "learning_rate": 9.90988039376942e-05,
1513
+ "loss": 0.0718,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 8.059701492537313,
1518
+ "grad_norm": 0.31339579820632935,
1519
+ "learning_rate": 9.90831111046988e-05,
1520
+ "loss": 0.0557,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 8.097014925373134,
1525
+ "grad_norm": 0.1968696266412735,
1526
+ "learning_rate": 9.90672840803519e-05,
1527
+ "loss": 0.0571,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 8.134328358208956,
1532
+ "grad_norm": 0.23931682109832764,
1533
+ "learning_rate": 9.905132290792394e-05,
1534
+ "loss": 0.0566,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 8.171641791044776,
1539
+ "grad_norm": 0.21741189062595367,
1540
+ "learning_rate": 9.903522763105218e-05,
1541
+ "loss": 0.0575,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 8.208955223880597,
1546
+ "grad_norm": 0.22874368727207184,
1547
+ "learning_rate": 9.901899829374047e-05,
1548
+ "loss": 0.0565,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 8.246268656716419,
1553
+ "grad_norm": 0.3441888093948364,
1554
+ "learning_rate": 9.900263494035921e-05,
1555
+ "loss": 0.0565,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 8.283582089552239,
1560
+ "grad_norm": 0.2539830803871155,
1561
+ "learning_rate": 9.89861376156452e-05,
1562
+ "loss": 0.0538,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 8.32089552238806,
1567
+ "grad_norm": 0.2235102653503418,
1568
+ "learning_rate": 9.896950636470147e-05,
1569
+ "loss": 0.0609,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 8.35820895522388,
1574
+ "grad_norm": 0.1941322684288025,
1575
+ "learning_rate": 9.895274123299723e-05,
1576
+ "loss": 0.0562,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 8.395522388059701,
1581
+ "grad_norm": 0.2691369950771332,
1582
+ "learning_rate": 9.893584226636772e-05,
1583
+ "loss": 0.0608,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 8.432835820895523,
1588
+ "grad_norm": 0.24730461835861206,
1589
+ "learning_rate": 9.891880951101407e-05,
1590
+ "loss": 0.0582,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 8.470149253731343,
1595
+ "grad_norm": 0.34785839915275574,
1596
+ "learning_rate": 9.890164301350318e-05,
1597
+ "loss": 0.0506,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 8.507462686567164,
1602
+ "grad_norm": 0.3625825345516205,
1603
+ "learning_rate": 9.888434282076758e-05,
1604
+ "loss": 0.0614,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 8.544776119402986,
1609
+ "grad_norm": 0.25210148096084595,
1610
+ "learning_rate": 9.886690898010535e-05,
1611
+ "loss": 0.0611,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 8.582089552238806,
1616
+ "grad_norm": 0.27312466502189636,
1617
+ "learning_rate": 9.884934153917997e-05,
1618
+ "loss": 0.0537,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 8.619402985074627,
1623
+ "grad_norm": 0.314647912979126,
1624
+ "learning_rate": 9.883164054602012e-05,
1625
+ "loss": 0.0602,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 8.656716417910447,
1630
+ "grad_norm": 0.21531912684440613,
1631
+ "learning_rate": 9.881380604901964e-05,
1632
+ "loss": 0.0552,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 8.694029850746269,
1637
+ "grad_norm": 0.23920664191246033,
1638
+ "learning_rate": 9.879583809693738e-05,
1639
+ "loss": 0.0613,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 8.73134328358209,
1644
+ "grad_norm": 0.21864956617355347,
1645
+ "learning_rate": 9.877773673889701e-05,
1646
+ "loss": 0.0649,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 8.76865671641791,
1651
+ "grad_norm": 0.27523377537727356,
1652
+ "learning_rate": 9.8759502024387e-05,
1653
+ "loss": 0.0606,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 8.805970149253731,
1658
+ "grad_norm": 0.24805469810962677,
1659
+ "learning_rate": 9.87411340032603e-05,
1660
+ "loss": 0.0549,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 8.843283582089553,
1665
+ "grad_norm": 0.23070092499256134,
1666
+ "learning_rate": 9.872263272573443e-05,
1667
+ "loss": 0.0562,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 8.880597014925373,
1672
+ "grad_norm": 0.20833946764469147,
1673
+ "learning_rate": 9.870399824239117e-05,
1674
+ "loss": 0.05,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 8.917910447761194,
1679
+ "grad_norm": 0.34507372975349426,
1680
+ "learning_rate": 9.868523060417646e-05,
1681
+ "loss": 0.0613,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 8.955223880597014,
1686
+ "grad_norm": 0.32865110039711,
1687
+ "learning_rate": 9.86663298624003e-05,
1688
+ "loss": 0.0621,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 8.992537313432836,
1693
+ "grad_norm": 0.21305270493030548,
1694
+ "learning_rate": 9.864729606873663e-05,
1695
+ "loss": 0.0572,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 9.029850746268657,
1700
+ "grad_norm": 0.28193730115890503,
1701
+ "learning_rate": 9.862812927522309e-05,
1702
+ "loss": 0.0555,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 9.067164179104477,
1707
+ "grad_norm": 0.3953789472579956,
1708
+ "learning_rate": 9.860882953426099e-05,
1709
+ "loss": 0.0536,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 9.104477611940299,
1714
+ "grad_norm": 0.23013322055339813,
1715
+ "learning_rate": 9.858939689861506e-05,
1716
+ "loss": 0.0572,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 9.14179104477612,
1721
+ "grad_norm": 0.2906680107116699,
1722
+ "learning_rate": 9.856983142141339e-05,
1723
+ "loss": 0.0592,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 9.17910447761194,
1728
+ "grad_norm": 0.23490828275680542,
1729
+ "learning_rate": 9.855013315614725e-05,
1730
+ "loss": 0.0583,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 9.216417910447761,
1735
+ "grad_norm": 0.22825880348682404,
1736
+ "learning_rate": 9.853030215667093e-05,
1737
+ "loss": 0.059,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 9.253731343283581,
1742
+ "grad_norm": 0.25871285796165466,
1743
+ "learning_rate": 9.851033847720166e-05,
1744
+ "loss": 0.0555,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 9.291044776119403,
1749
+ "grad_norm": 0.27220776677131653,
1750
+ "learning_rate": 9.849024217231935e-05,
1751
+ "loss": 0.0542,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 9.328358208955224,
1756
+ "grad_norm": 0.26534005999565125,
1757
+ "learning_rate": 9.847001329696653e-05,
1758
+ "loss": 0.0526,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 9.365671641791044,
1763
+ "grad_norm": 0.33486032485961914,
1764
+ "learning_rate": 9.844965190644817e-05,
1765
+ "loss": 0.0563,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 9.402985074626866,
1770
+ "grad_norm": 0.2949483394622803,
1771
+ "learning_rate": 9.842915805643155e-05,
1772
+ "loss": 0.0556,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 9.440298507462687,
1777
+ "grad_norm": 0.24123981595039368,
1778
+ "learning_rate": 9.840853180294608e-05,
1779
+ "loss": 0.05,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 9.477611940298507,
1784
+ "grad_norm": 0.22536049783229828,
1785
+ "learning_rate": 9.838777320238312e-05,
1786
+ "loss": 0.0522,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 9.514925373134329,
1791
+ "grad_norm": 0.23206663131713867,
1792
+ "learning_rate": 9.836688231149592e-05,
1793
+ "loss": 0.0591,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 9.552238805970148,
1798
+ "grad_norm": 0.28573134541511536,
1799
+ "learning_rate": 9.834585918739936e-05,
1800
+ "loss": 0.0568,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 9.58955223880597,
1805
+ "grad_norm": 0.2628820538520813,
1806
+ "learning_rate": 9.832470388756987e-05,
1807
+ "loss": 0.0571,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 9.626865671641792,
1812
+ "grad_norm": 0.2880440652370453,
1813
+ "learning_rate": 9.830341646984521e-05,
1814
+ "loss": 0.0559,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 9.664179104477611,
1819
+ "grad_norm": 0.1786259263753891,
1820
+ "learning_rate": 9.82819969924244e-05,
1821
+ "loss": 0.058,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 9.701492537313433,
1826
+ "grad_norm": 0.3501608073711395,
1827
+ "learning_rate": 9.826044551386744e-05,
1828
+ "loss": 0.0523,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 9.738805970149254,
1833
+ "grad_norm": 0.24757252633571625,
1834
+ "learning_rate": 9.823876209309527e-05,
1835
+ "loss": 0.0587,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 9.776119402985074,
1840
+ "grad_norm": 0.2556290626525879,
1841
+ "learning_rate": 9.821694678938953e-05,
1842
+ "loss": 0.0555,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 9.813432835820896,
1847
+ "grad_norm": 0.2561217248439789,
1848
+ "learning_rate": 9.819499966239243e-05,
1849
+ "loss": 0.052,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 9.850746268656717,
1854
+ "grad_norm": 0.2776634097099304,
1855
+ "learning_rate": 9.817292077210659e-05,
1856
+ "loss": 0.0498,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 9.888059701492537,
1861
+ "grad_norm": 0.20668549835681915,
1862
+ "learning_rate": 9.815071017889482e-05,
1863
+ "loss": 0.0517,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 9.925373134328359,
1868
+ "grad_norm": 0.3100263178348541,
1869
+ "learning_rate": 9.812836794348004e-05,
1870
+ "loss": 0.0633,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 9.962686567164178,
1875
+ "grad_norm": 0.2780782878398895,
1876
+ "learning_rate": 9.81058941269451e-05,
1877
+ "loss": 0.0581,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 10.0,
1882
+ "grad_norm": 0.28903728723526,
1883
+ "learning_rate": 9.808328879073251e-05,
1884
+ "loss": 0.0538,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 10.037313432835822,
1889
+ "grad_norm": 0.22727562487125397,
1890
+ "learning_rate": 9.806055199664446e-05,
1891
+ "loss": 0.0491,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 10.074626865671641,
1896
+ "grad_norm": 0.267918199300766,
1897
+ "learning_rate": 9.803768380684242e-05,
1898
+ "loss": 0.0562,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 10.111940298507463,
1903
+ "grad_norm": 0.2988606095314026,
1904
+ "learning_rate": 9.801468428384716e-05,
1905
+ "loss": 0.0566,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 10.149253731343283,
1910
+ "grad_norm": 0.2710281312465668,
1911
+ "learning_rate": 9.799155349053851e-05,
1912
+ "loss": 0.0541,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 10.186567164179104,
1917
+ "grad_norm": 0.15320520102977753,
1918
+ "learning_rate": 9.796829149015517e-05,
1919
+ "loss": 0.0548,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 10.223880597014926,
1924
+ "grad_norm": 0.2653089463710785,
1925
+ "learning_rate": 9.794489834629455e-05,
1926
+ "loss": 0.0599,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 10.261194029850746,
1931
+ "grad_norm": 0.19223959743976593,
1932
+ "learning_rate": 9.792137412291265e-05,
1933
+ "loss": 0.0494,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 10.298507462686567,
1938
+ "grad_norm": 0.20455987751483917,
1939
+ "learning_rate": 9.789771888432375e-05,
1940
+ "loss": 0.0538,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 10.335820895522389,
1945
+ "grad_norm": 0.24908749759197235,
1946
+ "learning_rate": 9.787393269520039e-05,
1947
+ "loss": 0.0481,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 10.373134328358208,
1952
+ "grad_norm": 0.3131813406944275,
1953
+ "learning_rate": 9.785001562057309e-05,
1954
+ "loss": 0.0526,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 10.41044776119403,
1959
+ "grad_norm": 0.24828971922397614,
1960
+ "learning_rate": 9.782596772583026e-05,
1961
+ "loss": 0.0489,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 10.447761194029852,
1966
+ "grad_norm": 0.21727119386196136,
1967
+ "learning_rate": 9.780178907671789e-05,
1968
+ "loss": 0.0532,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 10.485074626865671,
1973
+ "grad_norm": 0.20279547572135925,
1974
+ "learning_rate": 9.777747973933948e-05,
1975
+ "loss": 0.0565,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 10.522388059701493,
1980
+ "grad_norm": 0.17726702988147736,
1981
+ "learning_rate": 9.775303978015585e-05,
1982
+ "loss": 0.0437,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 10.559701492537313,
1987
+ "grad_norm": 0.18961119651794434,
1988
+ "learning_rate": 9.772846926598491e-05,
1989
+ "loss": 0.0584,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 10.597014925373134,
1994
+ "grad_norm": 0.2498980015516281,
1995
+ "learning_rate": 9.77037682640015e-05,
1996
+ "loss": 0.0496,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 10.634328358208956,
2001
+ "grad_norm": 0.16978798806667328,
2002
+ "learning_rate": 9.767893684173721e-05,
2003
+ "loss": 0.0469,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 10.671641791044776,
2008
+ "grad_norm": 0.16128584742546082,
2009
+ "learning_rate": 9.765397506708023e-05,
2010
+ "loss": 0.0533,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 10.708955223880597,
2015
+ "grad_norm": 0.20463155210018158,
2016
+ "learning_rate": 9.762888300827507e-05,
2017
+ "loss": 0.0464,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 10.746268656716419,
2022
+ "grad_norm": 0.30601629614830017,
2023
+ "learning_rate": 9.760366073392246e-05,
2024
+ "loss": 0.0489,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 10.783582089552239,
2029
+ "grad_norm": 0.2730671763420105,
2030
+ "learning_rate": 9.757830831297914e-05,
2031
+ "loss": 0.0495,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 10.82089552238806,
2036
+ "grad_norm": 0.251432865858078,
2037
+ "learning_rate": 9.755282581475769e-05,
2038
+ "loss": 0.0549,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 10.85820895522388,
2043
+ "grad_norm": 0.26670166850090027,
2044
+ "learning_rate": 9.752721330892624e-05,
2045
+ "loss": 0.061,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 10.895522388059701,
2050
+ "grad_norm": 0.2965967655181885,
2051
+ "learning_rate": 9.750147086550844e-05,
2052
+ "loss": 0.0473,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 10.932835820895523,
2057
+ "grad_norm": 0.683840274810791,
2058
+ "learning_rate": 9.747559855488313e-05,
2059
+ "loss": 0.0509,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 10.970149253731343,
2064
+ "grad_norm": 0.25740495324134827,
2065
+ "learning_rate": 9.744959644778422e-05,
2066
+ "loss": 0.0515,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 11.007462686567164,
2071
+ "grad_norm": 0.2880542278289795,
2072
+ "learning_rate": 9.742346461530048e-05,
2073
+ "loss": 0.0482,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 11.044776119402986,
2078
+ "grad_norm": 0.45032551884651184,
2079
+ "learning_rate": 9.739720312887535e-05,
2080
+ "loss": 0.0557,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 11.082089552238806,
2085
+ "grad_norm": 0.2829900085926056,
2086
+ "learning_rate": 9.73708120603067e-05,
2087
+ "loss": 0.052,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 11.119402985074627,
2092
+ "grad_norm": 0.309597373008728,
2093
+ "learning_rate": 9.734429148174675e-05,
2094
+ "loss": 0.0541,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 11.156716417910447,
2099
+ "grad_norm": 0.2433389127254486,
2100
+ "learning_rate": 9.731764146570173e-05,
2101
+ "loss": 0.0482,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 11.194029850746269,
2106
+ "grad_norm": 0.24458132684230804,
2107
+ "learning_rate": 9.729086208503174e-05,
2108
+ "loss": 0.0505,
2109
+ "step": 3000
2110
+ },
2111
+ {
2112
+ "epoch": 11.23134328358209,
2113
+ "grad_norm": 0.2305087298154831,
2114
+ "learning_rate": 9.726395341295062e-05,
2115
+ "loss": 0.0504,
2116
+ "step": 3010
2117
+ },
2118
+ {
2119
+ "epoch": 11.26865671641791,
2120
+ "grad_norm": 0.18110457062721252,
2121
+ "learning_rate": 9.723691552302562e-05,
2122
+ "loss": 0.0575,
2123
+ "step": 3020
2124
+ },
2125
+ {
2126
+ "epoch": 11.305970149253731,
2127
+ "grad_norm": 0.20407621562480927,
2128
+ "learning_rate": 9.720974848917735e-05,
2129
+ "loss": 0.0494,
2130
+ "step": 3030
2131
+ },
2132
+ {
2133
+ "epoch": 11.343283582089553,
2134
+ "grad_norm": 0.25924697518348694,
2135
+ "learning_rate": 9.718245238567939e-05,
2136
+ "loss": 0.0472,
2137
+ "step": 3040
2138
+ },
2139
+ {
2140
+ "epoch": 11.380597014925373,
2141
+ "grad_norm": 0.23041822016239166,
2142
+ "learning_rate": 9.715502728715826e-05,
2143
+ "loss": 0.0481,
2144
+ "step": 3050
2145
+ },
2146
+ {
2147
+ "epoch": 11.417910447761194,
2148
+ "grad_norm": 0.25381171703338623,
2149
+ "learning_rate": 9.712747326859315e-05,
2150
+ "loss": 0.0543,
2151
+ "step": 3060
2152
+ },
2153
+ {
2154
+ "epoch": 11.455223880597014,
2155
+ "grad_norm": 0.18027640879154205,
2156
+ "learning_rate": 9.709979040531569e-05,
2157
+ "loss": 0.055,
2158
+ "step": 3070
2159
+ },
2160
+ {
2161
+ "epoch": 11.492537313432836,
2162
+ "grad_norm": 0.2954868674278259,
2163
+ "learning_rate": 9.707197877300974e-05,
2164
+ "loss": 0.0473,
2165
+ "step": 3080
2166
+ },
2167
+ {
2168
+ "epoch": 11.529850746268657,
2169
+ "grad_norm": 0.25323861837387085,
2170
+ "learning_rate": 9.704403844771128e-05,
2171
+ "loss": 0.0509,
2172
+ "step": 3090
2173
+ },
2174
+ {
2175
+ "epoch": 11.567164179104477,
2176
+ "grad_norm": 0.36910176277160645,
2177
+ "learning_rate": 9.701596950580806e-05,
2178
+ "loss": 0.0504,
2179
+ "step": 3100
2180
+ },
2181
+ {
2182
+ "epoch": 11.604477611940299,
2183
+ "grad_norm": 0.34199246764183044,
2184
+ "learning_rate": 9.698777202403953e-05,
2185
+ "loss": 0.0526,
2186
+ "step": 3110
2187
+ },
2188
+ {
2189
+ "epoch": 11.64179104477612,
2190
+ "grad_norm": 0.2146557718515396,
2191
+ "learning_rate": 9.695944607949649e-05,
2192
+ "loss": 0.0579,
2193
+ "step": 3120
2194
+ },
2195
+ {
2196
+ "epoch": 11.67910447761194,
2197
+ "grad_norm": 0.20559175312519073,
2198
+ "learning_rate": 9.693099174962103e-05,
2199
+ "loss": 0.0514,
2200
+ "step": 3130
2201
+ },
2202
+ {
2203
+ "epoch": 11.716417910447761,
2204
+ "grad_norm": 0.2689419090747833,
2205
+ "learning_rate": 9.690240911220618e-05,
2206
+ "loss": 0.0534,
2207
+ "step": 3140
2208
+ },
2209
+ {
2210
+ "epoch": 11.753731343283581,
2211
+ "grad_norm": 0.34870603680610657,
2212
+ "learning_rate": 9.687369824539577e-05,
2213
+ "loss": 0.0485,
2214
+ "step": 3150
2215
+ },
2216
+ {
2217
+ "epoch": 11.791044776119403,
2218
+ "grad_norm": 0.15433363616466522,
2219
+ "learning_rate": 9.684485922768422e-05,
2220
+ "loss": 0.0418,
2221
+ "step": 3160
2222
+ },
2223
+ {
2224
+ "epoch": 11.828358208955224,
2225
+ "grad_norm": 0.26874423027038574,
2226
+ "learning_rate": 9.681589213791633e-05,
2227
+ "loss": 0.0537,
2228
+ "step": 3170
2229
+ },
2230
+ {
2231
+ "epoch": 11.865671641791044,
2232
+ "grad_norm": 0.3361654281616211,
2233
+ "learning_rate": 9.6786797055287e-05,
2234
+ "loss": 0.0474,
2235
+ "step": 3180
2236
+ },
2237
+ {
2238
+ "epoch": 11.902985074626866,
2239
+ "grad_norm": 0.17938771843910217,
2240
+ "learning_rate": 9.675757405934103e-05,
2241
+ "loss": 0.0443,
2242
+ "step": 3190
2243
+ },
2244
+ {
2245
+ "epoch": 11.940298507462687,
2246
+ "grad_norm": 0.31368622183799744,
2247
+ "learning_rate": 9.672822322997305e-05,
2248
+ "loss": 0.0594,
2249
+ "step": 3200
2250
+ },
2251
+ {
2252
+ "epoch": 11.977611940298507,
2253
+ "grad_norm": 0.16268151998519897,
2254
+ "learning_rate": 9.669874464742705e-05,
2255
+ "loss": 0.0487,
2256
+ "step": 3210
2257
+ },
2258
+ {
2259
+ "epoch": 12.014925373134329,
2260
+ "grad_norm": 0.23879969120025635,
2261
+ "learning_rate": 9.66691383922964e-05,
2262
+ "loss": 0.0484,
2263
+ "step": 3220
2264
+ },
2265
+ {
2266
+ "epoch": 12.052238805970148,
2267
+ "grad_norm": 0.2321789413690567,
2268
+ "learning_rate": 9.663940454552342e-05,
2269
+ "loss": 0.051,
2270
+ "step": 3230
2271
+ },
2272
+ {
2273
+ "epoch": 12.08955223880597,
2274
+ "grad_norm": 0.22873088717460632,
2275
+ "learning_rate": 9.660954318839933e-05,
2276
+ "loss": 0.0406,
2277
+ "step": 3240
2278
+ },
2279
+ {
2280
+ "epoch": 12.126865671641792,
2281
+ "grad_norm": 0.3767557740211487,
2282
+ "learning_rate": 9.657955440256395e-05,
2283
+ "loss": 0.0432,
2284
+ "step": 3250
2285
+ },
2286
+ {
2287
+ "epoch": 12.164179104477611,
2288
+ "grad_norm": 0.21569453179836273,
2289
+ "learning_rate": 9.654943827000548e-05,
2290
+ "loss": 0.0528,
2291
+ "step": 3260
2292
+ },
2293
+ {
2294
+ "epoch": 12.201492537313433,
2295
+ "grad_norm": 0.23698291182518005,
2296
+ "learning_rate": 9.651919487306025e-05,
2297
+ "loss": 0.0457,
2298
+ "step": 3270
2299
+ },
2300
+ {
2301
+ "epoch": 12.238805970149254,
2302
+ "grad_norm": 0.21086478233337402,
2303
+ "learning_rate": 9.648882429441257e-05,
2304
+ "loss": 0.0508,
2305
+ "step": 3280
2306
+ },
2307
+ {
2308
+ "epoch": 12.276119402985074,
2309
+ "grad_norm": 0.19763463735580444,
2310
+ "learning_rate": 9.645832661709444e-05,
2311
+ "loss": 0.0497,
2312
+ "step": 3290
2313
+ },
2314
+ {
2315
+ "epoch": 12.313432835820896,
2316
+ "grad_norm": 0.18413852155208588,
2317
+ "learning_rate": 9.642770192448536e-05,
2318
+ "loss": 0.0441,
2319
+ "step": 3300
2320
+ },
2321
+ {
2322
+ "epoch": 12.350746268656717,
2323
+ "grad_norm": 0.13946911692619324,
2324
+ "learning_rate": 9.639695030031204e-05,
2325
+ "loss": 0.0453,
2326
+ "step": 3310
2327
+ },
2328
+ {
2329
+ "epoch": 12.388059701492537,
2330
+ "grad_norm": 0.21613670885562897,
2331
+ "learning_rate": 9.636607182864827e-05,
2332
+ "loss": 0.0511,
2333
+ "step": 3320
2334
+ },
2335
+ {
2336
+ "epoch": 12.425373134328359,
2337
+ "grad_norm": 0.24953646957874298,
2338
+ "learning_rate": 9.63350665939146e-05,
2339
+ "loss": 0.0451,
2340
+ "step": 3330
2341
+ },
2342
+ {
2343
+ "epoch": 12.462686567164178,
2344
+ "grad_norm": 0.2993795871734619,
2345
+ "learning_rate": 9.630393468087818e-05,
2346
+ "loss": 0.0469,
2347
+ "step": 3340
2348
+ },
2349
+ {
2350
+ "epoch": 12.5,
2351
+ "grad_norm": 0.2261819839477539,
2352
+ "learning_rate": 9.627267617465243e-05,
2353
+ "loss": 0.0484,
2354
+ "step": 3350
2355
+ },
2356
+ {
2357
+ "epoch": 12.537313432835822,
2358
+ "grad_norm": 0.23026186227798462,
2359
+ "learning_rate": 9.624129116069694e-05,
2360
+ "loss": 0.0452,
2361
+ "step": 3360
2362
+ },
2363
+ {
2364
+ "epoch": 12.574626865671641,
2365
+ "grad_norm": 0.27859947085380554,
2366
+ "learning_rate": 9.620977972481716e-05,
2367
+ "loss": 0.0593,
2368
+ "step": 3370
2369
+ },
2370
+ {
2371
+ "epoch": 12.611940298507463,
2372
+ "grad_norm": 0.23060785233974457,
2373
+ "learning_rate": 9.617814195316411e-05,
2374
+ "loss": 0.05,
2375
+ "step": 3380
2376
+ },
2377
+ {
2378
+ "epoch": 12.649253731343283,
2379
+ "grad_norm": 0.20185025036334991,
2380
+ "learning_rate": 9.614637793223425e-05,
2381
+ "loss": 0.0573,
2382
+ "step": 3390
2383
+ },
2384
+ {
2385
+ "epoch": 12.686567164179104,
2386
+ "grad_norm": 0.3584498167037964,
2387
+ "learning_rate": 9.611448774886924e-05,
2388
+ "loss": 0.052,
2389
+ "step": 3400
2390
+ },
2391
+ {
2392
+ "epoch": 12.723880597014926,
2393
+ "grad_norm": 0.19336827099323273,
2394
+ "learning_rate": 9.60824714902556e-05,
2395
+ "loss": 0.0535,
2396
+ "step": 3410
2397
+ },
2398
+ {
2399
+ "epoch": 12.761194029850746,
2400
+ "grad_norm": 0.22223635017871857,
2401
+ "learning_rate": 9.605032924392457e-05,
2402
+ "loss": 0.05,
2403
+ "step": 3420
2404
+ },
2405
+ {
2406
+ "epoch": 12.798507462686567,
2407
+ "grad_norm": 0.17108851671218872,
2408
+ "learning_rate": 9.601806109775179e-05,
2409
+ "loss": 0.0475,
2410
+ "step": 3430
2411
+ },
2412
+ {
2413
+ "epoch": 12.835820895522389,
2414
+ "grad_norm": 0.3861902952194214,
2415
+ "learning_rate": 9.598566713995718e-05,
2416
+ "loss": 0.0439,
2417
+ "step": 3440
2418
+ },
2419
+ {
2420
+ "epoch": 12.873134328358208,
2421
+ "grad_norm": 0.18927253782749176,
2422
+ "learning_rate": 9.595314745910456e-05,
2423
+ "loss": 0.052,
2424
+ "step": 3450
2425
+ },
2426
+ {
2427
+ "epoch": 12.91044776119403,
2428
+ "grad_norm": 0.21963383257389069,
2429
+ "learning_rate": 9.59205021441015e-05,
2430
+ "loss": 0.0504,
2431
+ "step": 3460
2432
+ },
2433
+ {
2434
+ "epoch": 12.947761194029852,
2435
+ "grad_norm": 0.18016670644283295,
2436
+ "learning_rate": 9.588773128419906e-05,
2437
+ "loss": 0.0467,
2438
+ "step": 3470
2439
+ },
2440
+ {
2441
+ "epoch": 12.985074626865671,
2442
+ "grad_norm": 0.1776365041732788,
2443
+ "learning_rate": 9.58548349689915e-05,
2444
+ "loss": 0.0414,
2445
+ "step": 3480
2446
+ },
2447
+ {
2448
+ "epoch": 13.022388059701493,
2449
+ "grad_norm": 0.2616482973098755,
2450
+ "learning_rate": 9.582181328841611e-05,
2451
+ "loss": 0.0442,
2452
+ "step": 3490
2453
+ },
2454
+ {
2455
+ "epoch": 13.059701492537313,
2456
+ "grad_norm": 0.20341171324253082,
2457
+ "learning_rate": 9.578866633275288e-05,
2458
+ "loss": 0.0533,
2459
+ "step": 3500
2460
+ },
2461
+ {
2462
+ "epoch": 13.097014925373134,
2463
+ "grad_norm": 0.2223699688911438,
2464
+ "learning_rate": 9.575539419262434e-05,
2465
+ "loss": 0.0458,
2466
+ "step": 3510
2467
+ },
2468
+ {
2469
+ "epoch": 13.134328358208956,
2470
+ "grad_norm": 0.22557464241981506,
2471
+ "learning_rate": 9.572199695899522e-05,
2472
+ "loss": 0.0445,
2473
+ "step": 3520
2474
+ },
2475
+ {
2476
+ "epoch": 13.171641791044776,
2477
+ "grad_norm": 0.25104308128356934,
2478
+ "learning_rate": 9.568847472317232e-05,
2479
+ "loss": 0.0435,
2480
+ "step": 3530
2481
+ },
2482
+ {
2483
+ "epoch": 13.208955223880597,
2484
+ "grad_norm": 0.18720711767673492,
2485
+ "learning_rate": 9.565482757680415e-05,
2486
+ "loss": 0.0453,
2487
+ "step": 3540
2488
+ },
2489
+ {
2490
+ "epoch": 13.246268656716419,
2491
+ "grad_norm": 0.16838951408863068,
2492
+ "learning_rate": 9.562105561188069e-05,
2493
+ "loss": 0.0505,
2494
+ "step": 3550
2495
+ },
2496
+ {
2497
+ "epoch": 13.283582089552239,
2498
+ "grad_norm": 0.31681734323501587,
2499
+ "learning_rate": 9.558715892073323e-05,
2500
+ "loss": 0.0494,
2501
+ "step": 3560
2502
+ },
2503
+ {
2504
+ "epoch": 13.32089552238806,
2505
+ "grad_norm": 0.2390700727701187,
2506
+ "learning_rate": 9.555313759603402e-05,
2507
+ "loss": 0.0538,
2508
+ "step": 3570
2509
+ },
2510
+ {
2511
+ "epoch": 13.35820895522388,
2512
+ "grad_norm": 0.20680709183216095,
2513
+ "learning_rate": 9.551899173079607e-05,
2514
+ "loss": 0.0519,
2515
+ "step": 3580
2516
+ },
2517
+ {
2518
+ "epoch": 13.395522388059701,
2519
+ "grad_norm": 0.2758580148220062,
2520
+ "learning_rate": 9.548472141837286e-05,
2521
+ "loss": 0.0512,
2522
+ "step": 3590
2523
+ },
2524
+ {
2525
+ "epoch": 13.432835820895523,
2526
+ "grad_norm": 0.3653097450733185,
2527
+ "learning_rate": 9.545032675245813e-05,
2528
+ "loss": 0.0496,
2529
+ "step": 3600
2530
+ },
2531
+ {
2532
+ "epoch": 13.470149253731343,
2533
+ "grad_norm": 0.23886866867542267,
2534
+ "learning_rate": 9.541580782708557e-05,
2535
+ "loss": 0.0455,
2536
+ "step": 3610
2537
+ },
2538
+ {
2539
+ "epoch": 13.507462686567164,
2540
+ "grad_norm": 0.3280908465385437,
2541
+ "learning_rate": 9.538116473662861e-05,
2542
+ "loss": 0.0489,
2543
+ "step": 3620
2544
+ },
2545
+ {
2546
+ "epoch": 13.544776119402986,
2547
+ "grad_norm": 0.20268180966377258,
2548
+ "learning_rate": 9.534639757580013e-05,
2549
+ "loss": 0.0484,
2550
+ "step": 3630
2551
+ },
2552
+ {
2553
+ "epoch": 13.582089552238806,
2554
+ "grad_norm": 0.2582015097141266,
2555
+ "learning_rate": 9.531150643965223e-05,
2556
+ "loss": 0.0487,
2557
+ "step": 3640
2558
+ },
2559
+ {
2560
+ "epoch": 13.619402985074627,
2561
+ "grad_norm": 0.18157973885536194,
2562
+ "learning_rate": 9.527649142357596e-05,
2563
+ "loss": 0.0496,
2564
+ "step": 3650
2565
+ },
2566
+ {
2567
+ "epoch": 13.656716417910447,
2568
+ "grad_norm": 0.22841542959213257,
2569
+ "learning_rate": 9.524135262330098e-05,
2570
+ "loss": 0.0467,
2571
+ "step": 3660
2572
+ },
2573
+ {
2574
+ "epoch": 13.694029850746269,
2575
+ "grad_norm": 0.2519935369491577,
2576
+ "learning_rate": 9.520609013489547e-05,
2577
+ "loss": 0.0487,
2578
+ "step": 3670
2579
+ },
2580
+ {
2581
+ "epoch": 13.73134328358209,
2582
+ "grad_norm": 0.24680495262145996,
2583
+ "learning_rate": 9.517070405476575e-05,
2584
+ "loss": 0.0457,
2585
+ "step": 3680
2586
+ },
2587
+ {
2588
+ "epoch": 13.76865671641791,
2589
+ "grad_norm": 0.26362067461013794,
2590
+ "learning_rate": 9.513519447965595e-05,
2591
+ "loss": 0.0495,
2592
+ "step": 3690
2593
+ },
2594
+ {
2595
+ "epoch": 13.805970149253731,
2596
+ "grad_norm": 0.3240712583065033,
2597
+ "learning_rate": 9.509956150664796e-05,
2598
+ "loss": 0.0496,
2599
+ "step": 3700
2600
+ },
2601
+ {
2602
+ "epoch": 13.843283582089553,
2603
+ "grad_norm": 0.21009013056755066,
2604
+ "learning_rate": 9.50638052331609e-05,
2605
+ "loss": 0.0457,
2606
+ "step": 3710
2607
+ },
2608
+ {
2609
+ "epoch": 13.880597014925373,
2610
+ "grad_norm": 0.1669154316186905,
2611
+ "learning_rate": 9.502792575695112e-05,
2612
+ "loss": 0.0496,
2613
+ "step": 3720
2614
+ },
2615
+ {
2616
+ "epoch": 13.917910447761194,
2617
+ "grad_norm": 0.22347605228424072,
2618
+ "learning_rate": 9.499192317611167e-05,
2619
+ "loss": 0.0426,
2620
+ "step": 3730
2621
+ },
2622
+ {
2623
+ "epoch": 13.955223880597014,
2624
+ "grad_norm": 0.15208907425403595,
2625
+ "learning_rate": 9.49557975890723e-05,
2626
+ "loss": 0.0447,
2627
+ "step": 3740
2628
+ },
2629
+ {
2630
+ "epoch": 13.992537313432836,
2631
+ "grad_norm": 0.3206101059913635,
2632
+ "learning_rate": 9.491954909459895e-05,
2633
+ "loss": 0.0471,
2634
+ "step": 3750
2635
+ },
2636
+ {
2637
+ "epoch": 14.029850746268657,
2638
+ "grad_norm": 0.15873713791370392,
2639
+ "learning_rate": 9.488317779179361e-05,
2640
+ "loss": 0.0401,
2641
+ "step": 3760
2642
+ },
2643
+ {
2644
+ "epoch": 14.067164179104477,
2645
+ "grad_norm": 0.19690357148647308,
2646
+ "learning_rate": 9.484668378009408e-05,
2647
+ "loss": 0.0491,
2648
+ "step": 3770
2649
+ },
2650
+ {
2651
+ "epoch": 14.104477611940299,
2652
+ "grad_norm": 0.3211113214492798,
2653
+ "learning_rate": 9.481006715927351e-05,
2654
+ "loss": 0.049,
2655
+ "step": 3780
2656
+ },
2657
+ {
2658
+ "epoch": 14.14179104477612,
2659
+ "grad_norm": 0.27657604217529297,
2660
+ "learning_rate": 9.477332802944044e-05,
2661
+ "loss": 0.0396,
2662
+ "step": 3790
2663
+ },
2664
+ {
2665
+ "epoch": 14.17910447761194,
2666
+ "grad_norm": 0.20194031298160553,
2667
+ "learning_rate": 9.473646649103818e-05,
2668
+ "loss": 0.0442,
2669
+ "step": 3800
2670
+ },
2671
+ {
2672
+ "epoch": 14.216417910447761,
2673
+ "grad_norm": 0.20344595611095428,
2674
+ "learning_rate": 9.46994826448448e-05,
2675
+ "loss": 0.0427,
2676
+ "step": 3810
2677
+ },
2678
+ {
2679
+ "epoch": 14.253731343283581,
2680
+ "grad_norm": 0.2067718505859375,
2681
+ "learning_rate": 9.46623765919727e-05,
2682
+ "loss": 0.0501,
2683
+ "step": 3820
2684
+ },
2685
+ {
2686
+ "epoch": 14.291044776119403,
2687
+ "grad_norm": 0.29719170928001404,
2688
+ "learning_rate": 9.462514843386845e-05,
2689
+ "loss": 0.0519,
2690
+ "step": 3830
2691
+ },
2692
+ {
2693
+ "epoch": 14.328358208955224,
2694
+ "grad_norm": 0.2347182184457779,
2695
+ "learning_rate": 9.458779827231237e-05,
2696
+ "loss": 0.0413,
2697
+ "step": 3840
2698
+ },
2699
+ {
2700
+ "epoch": 14.365671641791044,
2701
+ "grad_norm": 0.1558852344751358,
2702
+ "learning_rate": 9.45503262094184e-05,
2703
+ "loss": 0.0442,
2704
+ "step": 3850
2705
+ },
2706
+ {
2707
+ "epoch": 14.402985074626866,
2708
+ "grad_norm": 0.23085005581378937,
2709
+ "learning_rate": 9.451273234763371e-05,
2710
+ "loss": 0.047,
2711
+ "step": 3860
2712
+ },
2713
+ {
2714
+ "epoch": 14.440298507462687,
2715
+ "grad_norm": 0.1515151560306549,
2716
+ "learning_rate": 9.447501678973852e-05,
2717
+ "loss": 0.0481,
2718
+ "step": 3870
2719
+ },
2720
+ {
2721
+ "epoch": 14.477611940298507,
2722
+ "grad_norm": 0.1916729211807251,
2723
+ "learning_rate": 9.443717963884569e-05,
2724
+ "loss": 0.0474,
2725
+ "step": 3880
2726
+ },
2727
+ {
2728
+ "epoch": 14.514925373134329,
2729
+ "grad_norm": 0.2536492943763733,
2730
+ "learning_rate": 9.439922099840054e-05,
2731
+ "loss": 0.0382,
2732
+ "step": 3890
2733
+ },
2734
+ {
2735
+ "epoch": 14.552238805970148,
2736
+ "grad_norm": 0.1672086864709854,
2737
+ "learning_rate": 9.43611409721806e-05,
2738
+ "loss": 0.0497,
2739
+ "step": 3900
2740
+ },
2741
+ {
2742
+ "epoch": 14.58955223880597,
2743
+ "grad_norm": 0.3644237518310547,
2744
+ "learning_rate": 9.432293966429514e-05,
2745
+ "loss": 0.0444,
2746
+ "step": 3910
2747
+ },
2748
+ {
2749
+ "epoch": 14.626865671641792,
2750
+ "grad_norm": 0.20307251811027527,
2751
+ "learning_rate": 9.428461717918511e-05,
2752
+ "loss": 0.0452,
2753
+ "step": 3920
2754
+ },
2755
+ {
2756
+ "epoch": 14.664179104477611,
2757
+ "grad_norm": 0.20441733300685883,
2758
+ "learning_rate": 9.424617362162271e-05,
2759
+ "loss": 0.0454,
2760
+ "step": 3930
2761
+ },
2762
+ {
2763
+ "epoch": 14.701492537313433,
2764
+ "grad_norm": 0.26315611600875854,
2765
+ "learning_rate": 9.420760909671118e-05,
2766
+ "loss": 0.0486,
2767
+ "step": 3940
2768
+ },
2769
+ {
2770
+ "epoch": 14.738805970149254,
2771
+ "grad_norm": 0.1983092874288559,
2772
+ "learning_rate": 9.416892370988444e-05,
2773
+ "loss": 0.0483,
2774
+ "step": 3950
2775
+ },
2776
+ {
2777
+ "epoch": 14.776119402985074,
2778
+ "grad_norm": 0.18301443755626678,
2779
+ "learning_rate": 9.413011756690685e-05,
2780
+ "loss": 0.0456,
2781
+ "step": 3960
2782
+ },
2783
+ {
2784
+ "epoch": 14.813432835820896,
2785
+ "grad_norm": 0.2433597594499588,
2786
+ "learning_rate": 9.409119077387294e-05,
2787
+ "loss": 0.0463,
2788
+ "step": 3970
2789
+ },
2790
+ {
2791
+ "epoch": 14.850746268656717,
2792
+ "grad_norm": 0.27949392795562744,
2793
+ "learning_rate": 9.405214343720707e-05,
2794
+ "loss": 0.0412,
2795
+ "step": 3980
2796
+ },
2797
+ {
2798
+ "epoch": 14.888059701492537,
2799
+ "grad_norm": 0.22806599736213684,
2800
+ "learning_rate": 9.401297566366318e-05,
2801
+ "loss": 0.0448,
2802
+ "step": 3990
2803
+ },
2804
+ {
2805
+ "epoch": 14.925373134328359,
2806
+ "grad_norm": 0.25421562790870667,
2807
+ "learning_rate": 9.397368756032445e-05,
2808
+ "loss": 0.0426,
2809
+ "step": 4000
2810
+ },
2811
+ {
2812
+ "epoch": 14.962686567164178,
2813
+ "grad_norm": 0.2436474859714508,
2814
+ "learning_rate": 9.393427923460308e-05,
2815
+ "loss": 0.0474,
2816
+ "step": 4010
2817
+ },
2818
+ {
2819
+ "epoch": 15.0,
2820
+ "grad_norm": 0.3756405711174011,
2821
+ "learning_rate": 9.389475079423988e-05,
2822
+ "loss": 0.0438,
2823
+ "step": 4020
2824
+ },
2825
+ {
2826
+ "epoch": 15.037313432835822,
2827
+ "grad_norm": 0.25687697529792786,
2828
+ "learning_rate": 9.385510234730415e-05,
2829
+ "loss": 0.0435,
2830
+ "step": 4030
2831
+ },
2832
+ {
2833
+ "epoch": 15.074626865671641,
2834
+ "grad_norm": 0.17263716459274292,
2835
+ "learning_rate": 9.381533400219318e-05,
2836
+ "loss": 0.0455,
2837
+ "step": 4040
2838
+ },
2839
+ {
2840
+ "epoch": 15.111940298507463,
2841
+ "grad_norm": 0.2471216470003128,
2842
+ "learning_rate": 9.377544586763215e-05,
2843
+ "loss": 0.0429,
2844
+ "step": 4050
2845
+ },
2846
+ {
2847
+ "epoch": 15.149253731343283,
2848
+ "grad_norm": 0.20195460319519043,
2849
+ "learning_rate": 9.373543805267368e-05,
2850
+ "loss": 0.0432,
2851
+ "step": 4060
2852
+ },
2853
+ {
2854
+ "epoch": 15.186567164179104,
2855
+ "grad_norm": 0.1709851622581482,
2856
+ "learning_rate": 9.369531066669758e-05,
2857
+ "loss": 0.0477,
2858
+ "step": 4070
2859
+ },
2860
+ {
2861
+ "epoch": 15.223880597014926,
2862
+ "grad_norm": 0.23063932359218597,
2863
+ "learning_rate": 9.365506381941066e-05,
2864
+ "loss": 0.0379,
2865
+ "step": 4080
2866
+ },
2867
+ {
2868
+ "epoch": 15.261194029850746,
2869
+ "grad_norm": 0.3265426754951477,
2870
+ "learning_rate": 9.36146976208462e-05,
2871
+ "loss": 0.0435,
2872
+ "step": 4090
2873
+ },
2874
+ {
2875
+ "epoch": 15.298507462686567,
2876
+ "grad_norm": 0.26373934745788574,
2877
+ "learning_rate": 9.357421218136386e-05,
2878
+ "loss": 0.047,
2879
+ "step": 4100
2880
+ },
2881
+ {
2882
+ "epoch": 15.335820895522389,
2883
+ "grad_norm": 0.16861388087272644,
2884
+ "learning_rate": 9.353360761164931e-05,
2885
+ "loss": 0.0448,
2886
+ "step": 4110
2887
+ },
2888
+ {
2889
+ "epoch": 15.373134328358208,
2890
+ "grad_norm": 0.303790807723999,
2891
+ "learning_rate": 9.349288402271388e-05,
2892
+ "loss": 0.0396,
2893
+ "step": 4120
2894
+ },
2895
+ {
2896
+ "epoch": 15.41044776119403,
2897
+ "grad_norm": 0.1940719038248062,
2898
+ "learning_rate": 9.345204152589428e-05,
2899
+ "loss": 0.0474,
2900
+ "step": 4130
2901
+ },
2902
+ {
2903
+ "epoch": 15.447761194029852,
2904
+ "grad_norm": 0.34091615676879883,
2905
+ "learning_rate": 9.341108023285238e-05,
2906
+ "loss": 0.0424,
2907
+ "step": 4140
2908
+ },
2909
+ {
2910
+ "epoch": 15.485074626865671,
2911
+ "grad_norm": 0.27036693692207336,
2912
+ "learning_rate": 9.337000025557476e-05,
2913
+ "loss": 0.0482,
2914
+ "step": 4150
2915
+ },
2916
+ {
2917
+ "epoch": 15.522388059701493,
2918
+ "grad_norm": 0.16908007860183716,
2919
+ "learning_rate": 9.332880170637252e-05,
2920
+ "loss": 0.0381,
2921
+ "step": 4160
2922
+ },
2923
+ {
2924
+ "epoch": 15.559701492537313,
2925
+ "grad_norm": 0.23332923650741577,
2926
+ "learning_rate": 9.328748469788093e-05,
2927
+ "loss": 0.0427,
2928
+ "step": 4170
2929
+ },
2930
+ {
2931
+ "epoch": 15.597014925373134,
2932
+ "grad_norm": 0.16899706423282623,
2933
+ "learning_rate": 9.32460493430591e-05,
2934
+ "loss": 0.0439,
2935
+ "step": 4180
2936
+ },
2937
+ {
2938
+ "epoch": 15.634328358208956,
2939
+ "grad_norm": 0.12869524955749512,
2940
+ "learning_rate": 9.320449575518972e-05,
2941
+ "loss": 0.0481,
2942
+ "step": 4190
2943
+ },
2944
+ {
2945
+ "epoch": 15.671641791044776,
2946
+ "grad_norm": 0.21159130334854126,
2947
+ "learning_rate": 9.316282404787871e-05,
2948
+ "loss": 0.0446,
2949
+ "step": 4200
2950
+ },
2951
+ {
2952
+ "epoch": 15.708955223880597,
2953
+ "grad_norm": 0.1849961131811142,
2954
+ "learning_rate": 9.31210343350549e-05,
2955
+ "loss": 0.041,
2956
+ "step": 4210
2957
+ },
2958
+ {
2959
+ "epoch": 15.746268656716419,
2960
+ "grad_norm": 0.16107840836048126,
2961
+ "learning_rate": 9.30791267309698e-05,
2962
+ "loss": 0.0429,
2963
+ "step": 4220
2964
+ },
2965
+ {
2966
+ "epoch": 15.783582089552239,
2967
+ "grad_norm": 0.14206446707248688,
2968
+ "learning_rate": 9.30371013501972e-05,
2969
+ "loss": 0.0409,
2970
+ "step": 4230
2971
+ },
2972
+ {
2973
+ "epoch": 15.82089552238806,
2974
+ "grad_norm": 0.2168441116809845,
2975
+ "learning_rate": 9.299495830763286e-05,
2976
+ "loss": 0.0413,
2977
+ "step": 4240
2978
+ },
2979
+ {
2980
+ "epoch": 15.85820895522388,
2981
+ "grad_norm": 0.21431951224803925,
2982
+ "learning_rate": 9.295269771849427e-05,
2983
+ "loss": 0.0472,
2984
+ "step": 4250
2985
+ },
2986
+ {
2987
+ "epoch": 15.895522388059701,
2988
+ "grad_norm": 0.16851255297660828,
2989
+ "learning_rate": 9.291031969832026e-05,
2990
+ "loss": 0.0508,
2991
+ "step": 4260
2992
+ },
2993
+ {
2994
+ "epoch": 15.932835820895523,
2995
+ "grad_norm": 0.18404732644557953,
2996
+ "learning_rate": 9.286782436297073e-05,
2997
+ "loss": 0.0402,
2998
+ "step": 4270
2999
+ },
3000
+ {
3001
+ "epoch": 15.970149253731343,
3002
+ "grad_norm": 0.21722930669784546,
3003
+ "learning_rate": 9.282521182862629e-05,
3004
+ "loss": 0.0397,
3005
+ "step": 4280
3006
+ },
3007
+ {
3008
+ "epoch": 16.007462686567163,
3009
+ "grad_norm": 0.2523709833621979,
3010
+ "learning_rate": 9.278248221178798e-05,
3011
+ "loss": 0.0427,
3012
+ "step": 4290
3013
+ },
3014
+ {
3015
+ "epoch": 16.044776119402986,
3016
+ "grad_norm": 0.17736563086509705,
3017
+ "learning_rate": 9.273963562927695e-05,
3018
+ "loss": 0.0458,
3019
+ "step": 4300
3020
+ },
3021
+ {
3022
+ "epoch": 16.082089552238806,
3023
+ "grad_norm": 0.20613858103752136,
3024
+ "learning_rate": 9.269667219823412e-05,
3025
+ "loss": 0.0387,
3026
+ "step": 4310
3027
+ },
3028
+ {
3029
+ "epoch": 16.119402985074625,
3030
+ "grad_norm": 0.16557513177394867,
3031
+ "learning_rate": 9.265359203611987e-05,
3032
+ "loss": 0.0411,
3033
+ "step": 4320
3034
+ },
3035
+ {
3036
+ "epoch": 16.15671641791045,
3037
+ "grad_norm": 0.28119519352912903,
3038
+ "learning_rate": 9.261039526071374e-05,
3039
+ "loss": 0.0468,
3040
+ "step": 4330
3041
+ },
3042
+ {
3043
+ "epoch": 16.19402985074627,
3044
+ "grad_norm": 0.21538576483726501,
3045
+ "learning_rate": 9.256708199011401e-05,
3046
+ "loss": 0.0368,
3047
+ "step": 4340
3048
+ },
3049
+ {
3050
+ "epoch": 16.23134328358209,
3051
+ "grad_norm": 0.19657357037067413,
3052
+ "learning_rate": 9.252365234273755e-05,
3053
+ "loss": 0.038,
3054
+ "step": 4350
3055
+ },
3056
+ {
3057
+ "epoch": 16.26865671641791,
3058
+ "grad_norm": 0.19258421659469604,
3059
+ "learning_rate": 9.248010643731935e-05,
3060
+ "loss": 0.0414,
3061
+ "step": 4360
3062
+ },
3063
+ {
3064
+ "epoch": 16.30597014925373,
3065
+ "grad_norm": 0.28801625967025757,
3066
+ "learning_rate": 9.243644439291223e-05,
3067
+ "loss": 0.0387,
3068
+ "step": 4370
3069
+ },
3070
+ {
3071
+ "epoch": 16.34328358208955,
3072
+ "grad_norm": 0.16581468284130096,
3073
+ "learning_rate": 9.239266632888659e-05,
3074
+ "loss": 0.0383,
3075
+ "step": 4380
3076
+ },
3077
+ {
3078
+ "epoch": 16.380597014925375,
3079
+ "grad_norm": 0.34664949774742126,
3080
+ "learning_rate": 9.234877236492997e-05,
3081
+ "loss": 0.0453,
3082
+ "step": 4390
3083
+ },
3084
+ {
3085
+ "epoch": 16.417910447761194,
3086
+ "grad_norm": 0.1439947783946991,
3087
+ "learning_rate": 9.230476262104677e-05,
3088
+ "loss": 0.0466,
3089
+ "step": 4400
3090
+ },
3091
+ {
3092
+ "epoch": 16.455223880597014,
3093
+ "grad_norm": 0.15509940683841705,
3094
+ "learning_rate": 9.226063721755799e-05,
3095
+ "loss": 0.0488,
3096
+ "step": 4410
3097
+ },
3098
+ {
3099
+ "epoch": 16.492537313432837,
3100
+ "grad_norm": 0.18005985021591187,
3101
+ "learning_rate": 9.221639627510076e-05,
3102
+ "loss": 0.0407,
3103
+ "step": 4420
3104
+ },
3105
+ {
3106
+ "epoch": 16.529850746268657,
3107
+ "grad_norm": 0.16012470424175262,
3108
+ "learning_rate": 9.217203991462815e-05,
3109
+ "loss": 0.0394,
3110
+ "step": 4430
3111
+ },
3112
+ {
3113
+ "epoch": 16.567164179104477,
3114
+ "grad_norm": 0.2978847920894623,
3115
+ "learning_rate": 9.212756825740873e-05,
3116
+ "loss": 0.0451,
3117
+ "step": 4440
3118
+ },
3119
+ {
3120
+ "epoch": 16.604477611940297,
3121
+ "grad_norm": 0.2236834019422531,
3122
+ "learning_rate": 9.208298142502636e-05,
3123
+ "loss": 0.0487,
3124
+ "step": 4450
3125
+ },
3126
+ {
3127
+ "epoch": 16.64179104477612,
3128
+ "grad_norm": 0.2686060667037964,
3129
+ "learning_rate": 9.20382795393797e-05,
3130
+ "loss": 0.0403,
3131
+ "step": 4460
3132
+ },
3133
+ {
3134
+ "epoch": 16.67910447761194,
3135
+ "grad_norm": 0.33534038066864014,
3136
+ "learning_rate": 9.199346272268199e-05,
3137
+ "loss": 0.0385,
3138
+ "step": 4470
3139
+ },
3140
+ {
3141
+ "epoch": 16.71641791044776,
3142
+ "grad_norm": 0.19250528514385223,
3143
+ "learning_rate": 9.194853109746074e-05,
3144
+ "loss": 0.0441,
3145
+ "step": 4480
3146
+ },
3147
+ {
3148
+ "epoch": 16.753731343283583,
3149
+ "grad_norm": 0.19218407571315765,
3150
+ "learning_rate": 9.190348478655724e-05,
3151
+ "loss": 0.0474,
3152
+ "step": 4490
3153
+ },
3154
+ {
3155
+ "epoch": 16.791044776119403,
3156
+ "grad_norm": 0.21163488924503326,
3157
+ "learning_rate": 9.185832391312644e-05,
3158
+ "loss": 0.0411,
3159
+ "step": 4500
3160
+ },
3161
+ {
3162
+ "epoch": 16.828358208955223,
3163
+ "grad_norm": 0.1758819818496704,
3164
+ "learning_rate": 9.18130486006364e-05,
3165
+ "loss": 0.0462,
3166
+ "step": 4510
3167
+ },
3168
+ {
3169
+ "epoch": 16.865671641791046,
3170
+ "grad_norm": 0.18571069836616516,
3171
+ "learning_rate": 9.176765897286813e-05,
3172
+ "loss": 0.0425,
3173
+ "step": 4520
3174
+ },
3175
+ {
3176
+ "epoch": 16.902985074626866,
3177
+ "grad_norm": 0.20819155871868134,
3178
+ "learning_rate": 9.17221551539151e-05,
3179
+ "loss": 0.0428,
3180
+ "step": 4530
3181
+ },
3182
+ {
3183
+ "epoch": 16.940298507462686,
3184
+ "grad_norm": 0.30357328057289124,
3185
+ "learning_rate": 9.167653726818305e-05,
3186
+ "loss": 0.0414,
3187
+ "step": 4540
3188
+ },
3189
+ {
3190
+ "epoch": 16.97761194029851,
3191
+ "grad_norm": 0.20977462828159332,
3192
+ "learning_rate": 9.163080544038952e-05,
3193
+ "loss": 0.0447,
3194
+ "step": 4550
3195
+ },
3196
+ {
3197
+ "epoch": 17.01492537313433,
3198
+ "grad_norm": 0.2535971701145172,
3199
+ "learning_rate": 9.158495979556358e-05,
3200
+ "loss": 0.0384,
3201
+ "step": 4560
3202
+ },
3203
+ {
3204
+ "epoch": 17.05223880597015,
3205
+ "grad_norm": 0.2789897620677948,
3206
+ "learning_rate": 9.153900045904549e-05,
3207
+ "loss": 0.042,
3208
+ "step": 4570
3209
+ },
3210
+ {
3211
+ "epoch": 17.08955223880597,
3212
+ "grad_norm": 0.18474848568439484,
3213
+ "learning_rate": 9.14929275564863e-05,
3214
+ "loss": 0.0398,
3215
+ "step": 4580
3216
+ },
3217
+ {
3218
+ "epoch": 17.12686567164179,
3219
+ "grad_norm": 0.12615208327770233,
3220
+ "learning_rate": 9.144674121384757e-05,
3221
+ "loss": 0.0466,
3222
+ "step": 4590
3223
+ },
3224
+ {
3225
+ "epoch": 17.16417910447761,
3226
+ "grad_norm": 0.17756640911102295,
3227
+ "learning_rate": 9.140044155740101e-05,
3228
+ "loss": 0.035,
3229
+ "step": 4600
3230
+ },
3231
+ {
3232
+ "epoch": 17.20149253731343,
3233
+ "grad_norm": 0.24410821497440338,
3234
+ "learning_rate": 9.135402871372808e-05,
3235
+ "loss": 0.0459,
3236
+ "step": 4610
3237
+ },
3238
+ {
3239
+ "epoch": 17.238805970149254,
3240
+ "grad_norm": 0.21573011577129364,
3241
+ "learning_rate": 9.130750280971978e-05,
3242
+ "loss": 0.0385,
3243
+ "step": 4620
3244
+ },
3245
+ {
3246
+ "epoch": 17.276119402985074,
3247
+ "grad_norm": 0.13879653811454773,
3248
+ "learning_rate": 9.126086397257612e-05,
3249
+ "loss": 0.0391,
3250
+ "step": 4630
3251
+ },
3252
+ {
3253
+ "epoch": 17.313432835820894,
3254
+ "grad_norm": 0.17508305609226227,
3255
+ "learning_rate": 9.121411232980588e-05,
3256
+ "loss": 0.038,
3257
+ "step": 4640
3258
+ },
3259
+ {
3260
+ "epoch": 17.350746268656717,
3261
+ "grad_norm": 0.2536008358001709,
3262
+ "learning_rate": 9.116724800922629e-05,
3263
+ "loss": 0.0418,
3264
+ "step": 4650
3265
+ },
3266
+ {
3267
+ "epoch": 17.388059701492537,
3268
+ "grad_norm": 0.1942976713180542,
3269
+ "learning_rate": 9.112027113896262e-05,
3270
+ "loss": 0.052,
3271
+ "step": 4660
3272
+ },
3273
+ {
3274
+ "epoch": 17.425373134328357,
3275
+ "grad_norm": 0.16561119258403778,
3276
+ "learning_rate": 9.107318184744781e-05,
3277
+ "loss": 0.0451,
3278
+ "step": 4670
3279
+ },
3280
+ {
3281
+ "epoch": 17.46268656716418,
3282
+ "grad_norm": 0.22971832752227783,
3283
+ "learning_rate": 9.102598026342222e-05,
3284
+ "loss": 0.0407,
3285
+ "step": 4680
3286
+ },
3287
+ {
3288
+ "epoch": 17.5,
3289
+ "grad_norm": 0.1306753158569336,
3290
+ "learning_rate": 9.097866651593317e-05,
3291
+ "loss": 0.042,
3292
+ "step": 4690
3293
+ },
3294
+ {
3295
+ "epoch": 17.53731343283582,
3296
+ "grad_norm": 0.21278400719165802,
3297
+ "learning_rate": 9.093124073433463e-05,
3298
+ "loss": 0.0458,
3299
+ "step": 4700
3300
+ },
3301
+ {
3302
+ "epoch": 17.574626865671643,
3303
+ "grad_norm": 0.22757171094417572,
3304
+ "learning_rate": 9.088370304828685e-05,
3305
+ "loss": 0.0364,
3306
+ "step": 4710
3307
+ },
3308
+ {
3309
+ "epoch": 17.611940298507463,
3310
+ "grad_norm": 0.216596320271492,
3311
+ "learning_rate": 9.083605358775612e-05,
3312
+ "loss": 0.0434,
3313
+ "step": 4720
3314
+ },
3315
+ {
3316
+ "epoch": 17.649253731343283,
3317
+ "grad_norm": 0.13022471964359283,
3318
+ "learning_rate": 9.078829248301417e-05,
3319
+ "loss": 0.0415,
3320
+ "step": 4730
3321
+ },
3322
+ {
3323
+ "epoch": 17.686567164179106,
3324
+ "grad_norm": 0.2280716598033905,
3325
+ "learning_rate": 9.074041986463808e-05,
3326
+ "loss": 0.0385,
3327
+ "step": 4740
3328
+ },
3329
+ {
3330
+ "epoch": 17.723880597014926,
3331
+ "grad_norm": 0.14666135609149933,
3332
+ "learning_rate": 9.069243586350975e-05,
3333
+ "loss": 0.0347,
3334
+ "step": 4750
3335
+ },
3336
+ {
3337
+ "epoch": 17.761194029850746,
3338
+ "grad_norm": 0.1631281077861786,
3339
+ "learning_rate": 9.064434061081562e-05,
3340
+ "loss": 0.0407,
3341
+ "step": 4760
3342
+ },
3343
+ {
3344
+ "epoch": 17.798507462686565,
3345
+ "grad_norm": 0.18697327375411987,
3346
+ "learning_rate": 9.059613423804623e-05,
3347
+ "loss": 0.0425,
3348
+ "step": 4770
3349
+ },
3350
+ {
3351
+ "epoch": 17.83582089552239,
3352
+ "grad_norm": 0.12955111265182495,
3353
+ "learning_rate": 9.0547816876996e-05,
3354
+ "loss": 0.0417,
3355
+ "step": 4780
3356
+ },
3357
+ {
3358
+ "epoch": 17.87313432835821,
3359
+ "grad_norm": 0.15547148883342743,
3360
+ "learning_rate": 9.049938865976275e-05,
3361
+ "loss": 0.0409,
3362
+ "step": 4790
3363
+ },
3364
+ {
3365
+ "epoch": 17.91044776119403,
3366
+ "grad_norm": 0.1900598704814911,
3367
+ "learning_rate": 9.045084971874738e-05,
3368
+ "loss": 0.0369,
3369
+ "step": 4800
3370
+ },
3371
+ {
3372
+ "epoch": 17.94776119402985,
3373
+ "grad_norm": 0.1846715807914734,
3374
+ "learning_rate": 9.040220018665347e-05,
3375
+ "loss": 0.0415,
3376
+ "step": 4810
3377
+ },
3378
+ {
3379
+ "epoch": 17.98507462686567,
3380
+ "grad_norm": 0.1829937845468521,
3381
+ "learning_rate": 9.035344019648702e-05,
3382
+ "loss": 0.0407,
3383
+ "step": 4820
3384
+ },
3385
+ {
3386
+ "epoch": 18.02238805970149,
3387
+ "grad_norm": 0.25900354981422424,
3388
+ "learning_rate": 9.030456988155596e-05,
3389
+ "loss": 0.0398,
3390
+ "step": 4830
3391
+ },
3392
+ {
3393
+ "epoch": 18.059701492537314,
3394
+ "grad_norm": 0.21235992014408112,
3395
+ "learning_rate": 9.025558937546988e-05,
3396
+ "loss": 0.0477,
3397
+ "step": 4840
3398
+ },
3399
+ {
3400
+ "epoch": 18.097014925373134,
3401
+ "grad_norm": 0.18785078823566437,
3402
+ "learning_rate": 9.020649881213958e-05,
3403
+ "loss": 0.039,
3404
+ "step": 4850
3405
+ },
3406
+ {
3407
+ "epoch": 18.134328358208954,
3408
+ "grad_norm": 0.1951548010110855,
3409
+ "learning_rate": 9.015729832577681e-05,
3410
+ "loss": 0.0357,
3411
+ "step": 4860
3412
+ },
3413
+ {
3414
+ "epoch": 18.171641791044777,
3415
+ "grad_norm": 0.1280934363603592,
3416
+ "learning_rate": 9.010798805089384e-05,
3417
+ "loss": 0.0425,
3418
+ "step": 4870
3419
+ },
3420
+ {
3421
+ "epoch": 18.208955223880597,
3422
+ "grad_norm": 0.1693423092365265,
3423
+ "learning_rate": 9.005856812230304e-05,
3424
+ "loss": 0.0447,
3425
+ "step": 4880
3426
+ },
3427
+ {
3428
+ "epoch": 18.246268656716417,
3429
+ "grad_norm": 0.23712658882141113,
3430
+ "learning_rate": 9.000903867511666e-05,
3431
+ "loss": 0.042,
3432
+ "step": 4890
3433
+ },
3434
+ {
3435
+ "epoch": 18.28358208955224,
3436
+ "grad_norm": 0.26489710807800293,
3437
+ "learning_rate": 8.995939984474624e-05,
3438
+ "loss": 0.0457,
3439
+ "step": 4900
3440
+ },
3441
+ {
3442
+ "epoch": 18.32089552238806,
3443
+ "grad_norm": 0.20792756974697113,
3444
+ "learning_rate": 8.990965176690252e-05,
3445
+ "loss": 0.0422,
3446
+ "step": 4910
3447
+ },
3448
+ {
3449
+ "epoch": 18.35820895522388,
3450
+ "grad_norm": 0.18526089191436768,
3451
+ "learning_rate": 8.98597945775948e-05,
3452
+ "loss": 0.0366,
3453
+ "step": 4920
3454
+ },
3455
+ {
3456
+ "epoch": 18.395522388059703,
3457
+ "grad_norm": 0.2214607298374176,
3458
+ "learning_rate": 8.980982841313074e-05,
3459
+ "loss": 0.0405,
3460
+ "step": 4930
3461
+ },
3462
+ {
3463
+ "epoch": 18.432835820895523,
3464
+ "grad_norm": 0.1896953135728836,
3465
+ "learning_rate": 8.975975341011596e-05,
3466
+ "loss": 0.0391,
3467
+ "step": 4940
3468
+ },
3469
+ {
3470
+ "epoch": 18.470149253731343,
3471
+ "grad_norm": 0.1430232971906662,
3472
+ "learning_rate": 8.970956970545355e-05,
3473
+ "loss": 0.0403,
3474
+ "step": 4950
3475
+ },
3476
+ {
3477
+ "epoch": 18.507462686567163,
3478
+ "grad_norm": 0.1991272121667862,
3479
+ "learning_rate": 8.965927743634391e-05,
3480
+ "loss": 0.0429,
3481
+ "step": 4960
3482
+ },
3483
+ {
3484
+ "epoch": 18.544776119402986,
3485
+ "grad_norm": 0.2361849844455719,
3486
+ "learning_rate": 8.96088767402841e-05,
3487
+ "loss": 0.0416,
3488
+ "step": 4970
3489
+ },
3490
+ {
3491
+ "epoch": 18.582089552238806,
3492
+ "grad_norm": 0.25857019424438477,
3493
+ "learning_rate": 8.955836775506776e-05,
3494
+ "loss": 0.0461,
3495
+ "step": 4980
3496
+ },
3497
+ {
3498
+ "epoch": 18.619402985074625,
3499
+ "grad_norm": 0.12873682379722595,
3500
+ "learning_rate": 8.950775061878453e-05,
3501
+ "loss": 0.035,
3502
+ "step": 4990
3503
+ },
3504
+ {
3505
+ "epoch": 18.65671641791045,
3506
+ "grad_norm": 0.19786769151687622,
3507
+ "learning_rate": 8.945702546981969e-05,
3508
+ "loss": 0.0399,
3509
+ "step": 5000
3510
+ },
3511
+ {
3512
+ "epoch": 18.69402985074627,
3513
+ "grad_norm": 0.2562239170074463,
3514
+ "learning_rate": 8.940619244685388e-05,
3515
+ "loss": 0.0372,
3516
+ "step": 5010
3517
+ },
3518
+ {
3519
+ "epoch": 18.73134328358209,
3520
+ "grad_norm": 0.14586858451366425,
3521
+ "learning_rate": 8.935525168886262e-05,
3522
+ "loss": 0.0427,
3523
+ "step": 5020
3524
+ },
3525
+ {
3526
+ "epoch": 18.76865671641791,
3527
+ "grad_norm": 0.20062318444252014,
3528
+ "learning_rate": 8.930420333511606e-05,
3529
+ "loss": 0.0403,
3530
+ "step": 5030
3531
+ },
3532
+ {
3533
+ "epoch": 18.80597014925373,
3534
+ "grad_norm": 0.22698874771595,
3535
+ "learning_rate": 8.92530475251784e-05,
3536
+ "loss": 0.036,
3537
+ "step": 5040
3538
+ },
3539
+ {
3540
+ "epoch": 18.84328358208955,
3541
+ "grad_norm": 0.2103697657585144,
3542
+ "learning_rate": 8.920178439890765e-05,
3543
+ "loss": 0.0431,
3544
+ "step": 5050
3545
+ },
3546
+ {
3547
+ "epoch": 18.880597014925375,
3548
+ "grad_norm": 0.16042308509349823,
3549
+ "learning_rate": 8.91504140964553e-05,
3550
+ "loss": 0.0388,
3551
+ "step": 5060
3552
+ },
3553
+ {
3554
+ "epoch": 18.917910447761194,
3555
+ "grad_norm": 0.16874109208583832,
3556
+ "learning_rate": 8.909893675826574e-05,
3557
+ "loss": 0.0388,
3558
+ "step": 5070
3559
+ },
3560
+ {
3561
+ "epoch": 18.955223880597014,
3562
+ "grad_norm": 0.15569192171096802,
3563
+ "learning_rate": 8.90473525250761e-05,
3564
+ "loss": 0.0353,
3565
+ "step": 5080
3566
+ },
3567
+ {
3568
+ "epoch": 18.992537313432837,
3569
+ "grad_norm": 0.16723507642745972,
3570
+ "learning_rate": 8.899566153791566e-05,
3571
+ "loss": 0.0443,
3572
+ "step": 5090
3573
+ },
3574
+ {
3575
+ "epoch": 19.029850746268657,
3576
+ "grad_norm": 0.23284228146076202,
3577
+ "learning_rate": 8.894386393810563e-05,
3578
+ "loss": 0.05,
3579
+ "step": 5100
3580
+ },
3581
+ {
3582
+ "epoch": 19.067164179104477,
3583
+ "grad_norm": 0.1621718853712082,
3584
+ "learning_rate": 8.889195986725865e-05,
3585
+ "loss": 0.0369,
3586
+ "step": 5110
3587
+ },
3588
+ {
3589
+ "epoch": 19.104477611940297,
3590
+ "grad_norm": 0.17522747814655304,
3591
+ "learning_rate": 8.883994946727849e-05,
3592
+ "loss": 0.0475,
3593
+ "step": 5120
3594
+ },
3595
+ {
3596
+ "epoch": 19.14179104477612,
3597
+ "grad_norm": 0.16110533475875854,
3598
+ "learning_rate": 8.878783288035957e-05,
3599
+ "loss": 0.0383,
3600
+ "step": 5130
3601
+ },
3602
+ {
3603
+ "epoch": 19.17910447761194,
3604
+ "grad_norm": 0.2574177086353302,
3605
+ "learning_rate": 8.873561024898668e-05,
3606
+ "loss": 0.0383,
3607
+ "step": 5140
3608
+ },
3609
+ {
3610
+ "epoch": 19.21641791044776,
3611
+ "grad_norm": 0.14560100436210632,
3612
+ "learning_rate": 8.868328171593448e-05,
3613
+ "loss": 0.037,
3614
+ "step": 5150
3615
+ },
3616
+ {
3617
+ "epoch": 19.253731343283583,
3618
+ "grad_norm": 0.14456631243228912,
3619
+ "learning_rate": 8.863084742426719e-05,
3620
+ "loss": 0.0423,
3621
+ "step": 5160
3622
+ },
3623
+ {
3624
+ "epoch": 19.291044776119403,
3625
+ "grad_norm": 0.1403595507144928,
3626
+ "learning_rate": 8.857830751733815e-05,
3627
+ "loss": 0.0327,
3628
+ "step": 5170
3629
+ },
3630
+ {
3631
+ "epoch": 19.328358208955223,
3632
+ "grad_norm": 0.18462564051151276,
3633
+ "learning_rate": 8.852566213878947e-05,
3634
+ "loss": 0.037,
3635
+ "step": 5180
3636
+ },
3637
+ {
3638
+ "epoch": 19.365671641791046,
3639
+ "grad_norm": 0.20725117623806,
3640
+ "learning_rate": 8.84729114325516e-05,
3641
+ "loss": 0.0376,
3642
+ "step": 5190
3643
+ },
3644
+ {
3645
+ "epoch": 19.402985074626866,
3646
+ "grad_norm": 0.17023132741451263,
3647
+ "learning_rate": 8.842005554284296e-05,
3648
+ "loss": 0.0467,
3649
+ "step": 5200
3650
+ },
3651
+ {
3652
+ "epoch": 19.440298507462686,
3653
+ "grad_norm": 0.31033241748809814,
3654
+ "learning_rate": 8.836709461416952e-05,
3655
+ "loss": 0.0425,
3656
+ "step": 5210
3657
+ },
3658
+ {
3659
+ "epoch": 19.47761194029851,
3660
+ "grad_norm": 0.14057482779026031,
3661
+ "learning_rate": 8.831402879132446e-05,
3662
+ "loss": 0.0432,
3663
+ "step": 5220
3664
+ },
3665
+ {
3666
+ "epoch": 19.51492537313433,
3667
+ "grad_norm": 0.23247437179088593,
3668
+ "learning_rate": 8.82608582193877e-05,
3669
+ "loss": 0.0396,
3670
+ "step": 5230
3671
+ },
3672
+ {
3673
+ "epoch": 19.55223880597015,
3674
+ "grad_norm": 0.1305907964706421,
3675
+ "learning_rate": 8.820758304372557e-05,
3676
+ "loss": 0.0389,
3677
+ "step": 5240
3678
+ },
3679
+ {
3680
+ "epoch": 19.58955223880597,
3681
+ "grad_norm": 0.17093417048454285,
3682
+ "learning_rate": 8.815420340999033e-05,
3683
+ "loss": 0.0347,
3684
+ "step": 5250
3685
+ },
3686
+ {
3687
+ "epoch": 19.62686567164179,
3688
+ "grad_norm": 0.24105240404605865,
3689
+ "learning_rate": 8.810071946411989e-05,
3690
+ "loss": 0.0392,
3691
+ "step": 5260
3692
+ },
3693
+ {
3694
+ "epoch": 19.66417910447761,
3695
+ "grad_norm": 0.2234315127134323,
3696
+ "learning_rate": 8.804713135233731e-05,
3697
+ "loss": 0.0403,
3698
+ "step": 5270
3699
+ },
3700
+ {
3701
+ "epoch": 19.701492537313435,
3702
+ "grad_norm": 0.16947844624519348,
3703
+ "learning_rate": 8.799343922115044e-05,
3704
+ "loss": 0.0368,
3705
+ "step": 5280
3706
+ },
3707
+ {
3708
+ "epoch": 19.738805970149254,
3709
+ "grad_norm": 0.26133742928504944,
3710
+ "learning_rate": 8.79396432173515e-05,
3711
+ "loss": 0.041,
3712
+ "step": 5290
3713
+ },
3714
+ {
3715
+ "epoch": 19.776119402985074,
3716
+ "grad_norm": 0.2099352777004242,
3717
+ "learning_rate": 8.788574348801675e-05,
3718
+ "loss": 0.0363,
3719
+ "step": 5300
3720
+ },
3721
+ {
3722
+ "epoch": 19.813432835820894,
3723
+ "grad_norm": 0.1662513017654419,
3724
+ "learning_rate": 8.783174018050594e-05,
3725
+ "loss": 0.0409,
3726
+ "step": 5310
3727
+ },
3728
+ {
3729
+ "epoch": 19.850746268656717,
3730
+ "grad_norm": 0.18933714926242828,
3731
+ "learning_rate": 8.77776334424621e-05,
3732
+ "loss": 0.0348,
3733
+ "step": 5320
3734
+ },
3735
+ {
3736
+ "epoch": 19.888059701492537,
3737
+ "grad_norm": 0.21673552691936493,
3738
+ "learning_rate": 8.772342342181095e-05,
3739
+ "loss": 0.037,
3740
+ "step": 5330
3741
+ },
3742
+ {
3743
+ "epoch": 19.925373134328357,
3744
+ "grad_norm": 0.13009892404079437,
3745
+ "learning_rate": 8.766911026676064e-05,
3746
+ "loss": 0.0386,
3747
+ "step": 5340
3748
+ },
3749
+ {
3750
+ "epoch": 19.96268656716418,
3751
+ "grad_norm": 0.1655230075120926,
3752
+ "learning_rate": 8.761469412580125e-05,
3753
+ "loss": 0.0404,
3754
+ "step": 5350
3755
+ },
3756
+ {
3757
+ "epoch": 20.0,
3758
+ "grad_norm": 0.2821272611618042,
3759
+ "learning_rate": 8.756017514770443e-05,
3760
+ "loss": 0.0441,
3761
+ "step": 5360
3762
+ },
3763
+ {
3764
+ "epoch": 20.03731343283582,
3765
+ "grad_norm": 0.1302652508020401,
3766
+ "learning_rate": 8.750555348152298e-05,
3767
+ "loss": 0.0389,
3768
+ "step": 5370
3769
+ },
3770
+ {
3771
+ "epoch": 20.074626865671643,
3772
+ "grad_norm": 0.13331563770771027,
3773
+ "learning_rate": 8.745082927659047e-05,
3774
+ "loss": 0.0393,
3775
+ "step": 5380
3776
+ },
3777
+ {
3778
+ "epoch": 20.111940298507463,
3779
+ "grad_norm": 0.244130939245224,
3780
+ "learning_rate": 8.739600268252078e-05,
3781
+ "loss": 0.0372,
3782
+ "step": 5390
3783
+ },
3784
+ {
3785
+ "epoch": 20.149253731343283,
3786
+ "grad_norm": 0.20429308712482452,
3787
+ "learning_rate": 8.73410738492077e-05,
3788
+ "loss": 0.0387,
3789
+ "step": 5400
3790
+ },
3791
+ {
3792
+ "epoch": 20.186567164179106,
3793
+ "grad_norm": 0.2954719364643097,
3794
+ "learning_rate": 8.728604292682459e-05,
3795
+ "loss": 0.0404,
3796
+ "step": 5410
3797
+ },
3798
+ {
3799
+ "epoch": 20.223880597014926,
3800
+ "grad_norm": 0.20438429713249207,
3801
+ "learning_rate": 8.723091006582389e-05,
3802
+ "loss": 0.0359,
3803
+ "step": 5420
3804
+ },
3805
+ {
3806
+ "epoch": 20.261194029850746,
3807
+ "grad_norm": 0.17289331555366516,
3808
+ "learning_rate": 8.717567541693673e-05,
3809
+ "loss": 0.0357,
3810
+ "step": 5430
3811
+ },
3812
+ {
3813
+ "epoch": 20.298507462686565,
3814
+ "grad_norm": 0.24367138743400574,
3815
+ "learning_rate": 8.71203391311725e-05,
3816
+ "loss": 0.0392,
3817
+ "step": 5440
3818
+ },
3819
+ {
3820
+ "epoch": 20.33582089552239,
3821
+ "grad_norm": 0.21900270879268646,
3822
+ "learning_rate": 8.706490135981855e-05,
3823
+ "loss": 0.0419,
3824
+ "step": 5450
3825
+ },
3826
+ {
3827
+ "epoch": 20.37313432835821,
3828
+ "grad_norm": 0.1526443362236023,
3829
+ "learning_rate": 8.700936225443959e-05,
3830
+ "loss": 0.0333,
3831
+ "step": 5460
3832
+ },
3833
+ {
3834
+ "epoch": 20.41044776119403,
3835
+ "grad_norm": 0.24582353234291077,
3836
+ "learning_rate": 8.695372196687743e-05,
3837
+ "loss": 0.0417,
3838
+ "step": 5470
3839
+ },
3840
+ {
3841
+ "epoch": 20.44776119402985,
3842
+ "grad_norm": 0.21462485194206238,
3843
+ "learning_rate": 8.689798064925049e-05,
3844
+ "loss": 0.0347,
3845
+ "step": 5480
3846
+ },
3847
+ {
3848
+ "epoch": 20.48507462686567,
3849
+ "grad_norm": 0.17611616849899292,
3850
+ "learning_rate": 8.684213845395339e-05,
3851
+ "loss": 0.0395,
3852
+ "step": 5490
3853
+ },
3854
+ {
3855
+ "epoch": 20.52238805970149,
3856
+ "grad_norm": 0.19724012911319733,
3857
+ "learning_rate": 8.678619553365659e-05,
3858
+ "loss": 0.0332,
3859
+ "step": 5500
3860
+ }
3861
+ ],
3862
+ "logging_steps": 10,
3863
+ "max_steps": 20000,
3864
+ "num_input_tokens_seen": 0,
3865
+ "num_train_epochs": 75,
3866
+ "save_steps": 500,
3867
+ "stateful_callbacks": {
3868
+ "TrainerControl": {
3869
+ "args": {
3870
+ "should_epoch_stop": false,
3871
+ "should_evaluate": false,
3872
+ "should_log": false,
3873
+ "should_save": true,
3874
+ "should_training_stop": false
3875
+ },
3876
+ "attributes": {}
3877
+ }
3878
+ },
3879
+ "total_flos": 4.134508587240192e+17,
3880
+ "train_batch_size": 32,
3881
+ "trial_name": null,
3882
+ "trial_params": null
3883
+ }
checkpoint-6000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-6000/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-6000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe7e637252ccbc2072f81752b8b8609968b5b00d64efc6b714296170cc67cc63
3
+ size 123328576
checkpoint-6000/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-6000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ba63daa5ec509f37eb8ad97d3a5972c35bea7a8dd0db7d466efd0d3f329b3c8
3
+ size 246824634
checkpoint-6000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a72b98cbbed0760e18ed8782e6594f386d25ba2964d1c4523fa752e78decb29
3
+ size 14244
checkpoint-6000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9495ce8c64325a54dd12b90b11af20d3a01df15d5fab3ae0abdaff10fb2ae7f
3
+ size 1064
checkpoint-6000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-6500/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-6500/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }
checkpoint-6500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef0e30937cf6ac0448dd499578f580b4575192c99686e3ba6b4ea14a17946583
3
+ size 123328576
checkpoint-6500/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,275 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 47.021484375,
8
+ 135.263671875,
9
+ 178.505859375,
10
+ 78.3984375,
11
+ 56.77734375
12
+ ],
13
+ "min": [
14
+ -25.576171875,
15
+ 46.93359375,
16
+ 89.736328125,
17
+ -30.41015625,
18
+ -77.607421875
19
+ ],
20
+ "mean": [
21
+ 7.780572414398193,
22
+ 121.54933166503906,
23
+ 145.44825744628906,
24
+ 26.051393508911133,
25
+ -12.748016357421875
26
+ ],
27
+ "std": [
28
+ 11.060831069946289,
29
+ 21.937597274780273,
30
+ 17.16187286376953,
31
+ 19.231945037841797,
32
+ 14.66512680053711
33
+ ],
34
+ "q01": [
35
+ -17.578125,
36
+ 58.0078125,
37
+ 97.998046875,
38
+ -13.447265625,
39
+ -39.9005859375
40
+ ],
41
+ "q99": [
42
+ 36.650390625,
43
+ 134.47265625,
44
+ 178.41796875,
45
+ 66.65009765625,
46
+ 40.166015625
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 52.22222137451172
52
+ ],
53
+ "min": [
54
+ -3.846153974533081
55
+ ],
56
+ "mean": [
57
+ 10.933439254760742
58
+ ],
59
+ "std": [
60
+ 15.509913444519043
61
+ ],
62
+ "q01": [
63
+ -3.846153974533081
64
+ ],
65
+ "q99": [
66
+ 51.02564239501953
67
+ ]
68
+ },
69
+ "mobile_base": {
70
+ "max": [
71
+ 75.42072296142578,
72
+ 276.7638244628906,
73
+ 93.75
74
+ ],
75
+ "min": [
76
+ -170.01620483398438,
77
+ -274.5497131347656,
78
+ -93.75
79
+ ],
80
+ "mean": [
81
+ -0.31241804361343384,
82
+ 58.99717712402344,
83
+ 2.4293017387390137
84
+ ],
85
+ "std": [
86
+ 10.56183910369873,
87
+ 119.39802551269531,
88
+ 22.590484619140625
89
+ ],
90
+ "q01": [
91
+ -33.65809627532959,
92
+ -265.6932678222656,
93
+ -72.849609375
94
+ ],
95
+ "q99": [
96
+ 30.679615020751953,
97
+ 270.1214904785156,
98
+ 90.234375
99
+ ]
100
+ }
101
+ },
102
+ "action": {
103
+ "single_arm": {
104
+ "max": [
105
+ 37.96875,
106
+ 135.087890625,
107
+ 179.384765625,
108
+ 78.837890625,
109
+ 57.392578125
110
+ ],
111
+ "min": [
112
+ -26.279296875,
113
+ 47.373046875,
114
+ 89.912109375,
115
+ -31.640625,
116
+ -77.16796875
117
+ ],
118
+ "mean": [
119
+ 8.038639068603516,
120
+ 122.76031494140625,
121
+ 145.15855407714844,
122
+ 26.28432846069336,
123
+ -13.195321083068848
124
+ ],
125
+ "std": [
126
+ 11.36032772064209,
127
+ 21.925451278686523,
128
+ 17.071842193603516,
129
+ 19.503877639770508,
130
+ 14.882487297058105
131
+ ],
132
+ "q01": [
133
+ -18.10546875,
134
+ 58.623046875,
135
+ 98.26171875,
136
+ -14.326171875,
137
+ -40.078125
138
+ ],
139
+ "q99": [
140
+ 37.44140625,
141
+ 135.087890625,
142
+ 179.296875,
143
+ 67.1484375,
144
+ 40.869140625
145
+ ]
146
+ },
147
+ "gripper": {
148
+ "max": [
149
+ 52.646484375
150
+ ],
151
+ "min": [
152
+ -10.72265625
153
+ ],
154
+ "mean": [
155
+ 4.366570949554443
156
+ ],
157
+ "std": [
158
+ 18.90865707397461
159
+ ],
160
+ "q01": [
161
+ -10.546875
162
+ ],
163
+ "q99": [
164
+ 51.767578125
165
+ ]
166
+ },
167
+ "mobile_base": {
168
+ "max": [
169
+ 230.0971221923828,
170
+ 265.6932678222656,
171
+ 90.0
172
+ ],
173
+ "min": [
174
+ -230.0971221923828,
175
+ -265.6932678222656,
176
+ -90.0
177
+ ],
178
+ "mean": [
179
+ -0.36507830023765564,
180
+ 60.13115310668945,
181
+ 2.5394127368927
182
+ ],
183
+ "std": [
184
+ 15.02155590057373,
185
+ 129.06507873535156,
186
+ 27.82071304321289
187
+ ],
188
+ "q01": [
189
+ -0.02556634694337845,
190
+ -265.6932678222656,
191
+ -90.0
192
+ ],
193
+ "q99": [
194
+ 0.02556634694337845,
195
+ 265.6932678222656,
196
+ 90.0
197
+ ]
198
+ }
199
+ }
200
+ },
201
+ "modalities": {
202
+ "video": {
203
+ "wrist": {
204
+ "resolution": [
205
+ 640,
206
+ 480
207
+ ],
208
+ "channels": 3,
209
+ "fps": 30.0
210
+ },
211
+ "front": {
212
+ "resolution": [
213
+ 640,
214
+ 480
215
+ ],
216
+ "channels": 3,
217
+ "fps": 30.0
218
+ }
219
+ },
220
+ "state": {
221
+ "single_arm": {
222
+ "absolute": true,
223
+ "rotation_type": null,
224
+ "shape": [
225
+ 5
226
+ ],
227
+ "continuous": true
228
+ },
229
+ "gripper": {
230
+ "absolute": true,
231
+ "rotation_type": null,
232
+ "shape": [
233
+ 1
234
+ ],
235
+ "continuous": true
236
+ },
237
+ "mobile_base": {
238
+ "absolute": true,
239
+ "rotation_type": null,
240
+ "shape": [
241
+ 3
242
+ ],
243
+ "continuous": true
244
+ }
245
+ },
246
+ "action": {
247
+ "single_arm": {
248
+ "absolute": true,
249
+ "rotation_type": null,
250
+ "shape": [
251
+ 5
252
+ ],
253
+ "continuous": true
254
+ },
255
+ "gripper": {
256
+ "absolute": true,
257
+ "rotation_type": null,
258
+ "shape": [
259
+ 1
260
+ ],
261
+ "continuous": true
262
+ },
263
+ "mobile_base": {
264
+ "absolute": true,
265
+ "rotation_type": null,
266
+ "shape": [
267
+ 3
268
+ ],
269
+ "continuous": true
270
+ }
271
+ }
272
+ },
273
+ "embodiment_tag": "new_embodiment"
274
+ }
275
+ }
checkpoint-6500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b48f8520ef9abf5e83329a356c9f345d1daabe3f4564eab738688ad9403d423f
3
+ size 246824634
checkpoint-6500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac183675168f49ce92cfa88a64cc68cf198aafa3214b49550c76b3c23dcb507e
3
+ size 14244
checkpoint-6500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29ee88f0f3f63ecfc8da4336b17f02a8ebc3c4ed4823c0ead962ba0f4237fc5e
3
+ size 1064
checkpoint-6500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-7000/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.14.0
checkpoint-7000/adapter_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/home/ibru/.cache/huggingface/hub/models--nvidia--GR00T-N1-2B/snapshots/32e1fd2507f7739fad443e6b449c8188e0e02fcb",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 128,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 64,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "q_proj",
27
+ "k_proj",
28
+ "to_k",
29
+ "to_q",
30
+ "v_proj",
31
+ "to_v"
32
+ ],
33
+ "task_type": "CAUSAL_LM",
34
+ "use_dora": false,
35
+ "use_rslora": false
36
+ }