Hugging Face's logo

heybaeheef
/
KU_SW_Academy

Safetensors
Model card Files
xet
Community
heybaeheef commited on
Commit
03d1e33
·
1 Parent(s): 68292a8

Keep LoRA adapter only, remove optimizer and tokenizer

Browse files
Files changed (9) hide show
  1. checkpoints/added_tokens.json +0 -28
  2. checkpoints/optimizer.pt +0 -3
  3. checkpoints/rng_state.pth +0 -3
  4. checkpoints/scheduler.pt +0 -3
  5. checkpoints/special_tokens_map.json +0 -31
  6. checkpoints/tokenizer.json +0 -3
  7. checkpoints/tokenizer_config.json +0 -239
  8. checkpoints/trainer_state.json +0 -2239
  9. checkpoints/training_args.bin +0 -3
checkpoints/added_tokens.json DELETED
@@ -1,28 +0,0 @@
1
- {
2
- "</think>": 151668,
3
- "</tool_call>": 151658,
4
- "</tool_response>": 151666,
5
- "<think>": 151667,
6
- "<tool_call>": 151657,
7
- "<tool_response>": 151665,
8
- "<|box_end|>": 151649,
9
- "<|box_start|>": 151648,
10
- "<|endoftext|>": 151643,
11
- "<|file_sep|>": 151664,
12
- "<|fim_middle|>": 151660,
13
- "<|fim_pad|>": 151662,
14
- "<|fim_prefix|>": 151659,
15
- "<|fim_suffix|>": 151661,
16
- "<|im_end|>": 151645,
17
- "<|im_start|>": 151644,
18
- "<|image_pad|>": 151655,
19
- "<|object_ref_end|>": 151647,
20
- "<|object_ref_start|>": 151646,
21
- "<|quad_end|>": 151651,
22
- "<|quad_start|>": 151650,
23
- "<|repo_name|>": 151663,
24
- "<|video_pad|>": 151656,
25
- "<|vision_end|>": 151653,
26
- "<|vision_pad|>": 151654,
27
- "<|vision_start|>": 151652
28
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoints/optimizer.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:e5467afcfac91b2814c1f2c0f9c09b9c8285e4cf6052cefb80a8d87111835c47
3
- size 349601867
 
 
 
 
checkpoints/rng_state.pth DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:09293bb49e73801439a1ab6491fd9d17316c230e182853de90bab3704b97f390
3
- size 14645
 
 
 
 
checkpoints/scheduler.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:6373daa4ac59cf2f70923abdb9c2ebfe86659c39fc0bb5939f23e95ad5441f9a
3
- size 1465
 
 
 
 
checkpoints/special_tokens_map.json DELETED
@@ -1,31 +0,0 @@
1
- {
2
- "additional_special_tokens": [
3
- "<|im_start|>",
4
- "<|im_end|>",
5
- "<|object_ref_start|>",
6
- "<|object_ref_end|>",
7
- "<|box_start|>",
8
- "<|box_end|>",
9
- "<|quad_start|>",
10
- "<|quad_end|>",
11
- "<|vision_start|>",
12
- "<|vision_end|>",
13
- "<|vision_pad|>",
14
- "<|image_pad|>",
15
- "<|video_pad|>"
16
- ],
17
- "eos_token": {
18
- "content": "<|im_end|>",
19
- "lstrip": false,
20
- "normalized": false,
21
- "rstrip": false,
22
- "single_word": false
23
- },
24
- "pad_token": {
25
- "content": "<|endoftext|>",
26
- "lstrip": false,
27
- "normalized": false,
28
- "rstrip": false,
29
- "single_word": false
30
- }
31
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoints/tokenizer.json DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:1898b39d0bc7babd03f2570363e639f71a5b7449ab810665f12a437b297904f3
3
- size 11422934
 
 
 
 
checkpoints/tokenizer_config.json DELETED
@@ -1,239 +0,0 @@
1
- {
2
- "add_bos_token": false,
3
- "add_prefix_space": false,
4
- "added_tokens_decoder": {
5
- "151643": {
6
- "content": "<|endoftext|>",
7
- "lstrip": false,
8
- "normalized": false,
9
- "rstrip": false,
10
- "single_word": false,
11
- "special": true
12
- },
13
- "151644": {
14
- "content": "<|im_start|>",
15
- "lstrip": false,
16
- "normalized": false,
17
- "rstrip": false,
18
- "single_word": false,
19
- "special": true
20
- },
21
- "151645": {
22
- "content": "<|im_end|>",
23
- "lstrip": false,
24
- "normalized": false,
25
- "rstrip": false,
26
- "single_word": false,
27
- "special": true
28
- },
29
- "151646": {
30
- "content": "<|object_ref_start|>",
31
- "lstrip": false,
32
- "normalized": false,
33
- "rstrip": false,
34
- "single_word": false,
35
- "special": true
36
- },
37
- "151647": {
38
- "content": "<|object_ref_end|>",
39
- "lstrip": false,
40
- "normalized": false,
41
- "rstrip": false,
42
- "single_word": false,
43
- "special": true
44
- },
45
- "151648": {
46
- "content": "<|box_start|>",
47
- "lstrip": false,
48
- "normalized": false,
49
- "rstrip": false,
50
- "single_word": false,
51
- "special": true
52
- },
53
- "151649": {
54
- "content": "<|box_end|>",
55
- "lstrip": false,
56
- "normalized": false,
57
- "rstrip": false,
58
- "single_word": false,
59
- "special": true
60
- },
61
- "151650": {
62
- "content": "<|quad_start|>",
63
- "lstrip": false,
64
- "normalized": false,
65
- "rstrip": false,
66
- "single_word": false,
67
- "special": true
68
- },
69
- "151651": {
70
- "content": "<|quad_end|>",
71
- "lstrip": false,
72
- "normalized": false,
73
- "rstrip": false,
74
- "single_word": false,
75
- "special": true
76
- },
77
- "151652": {
78
- "content": "<|vision_start|>",
79
- "lstrip": false,
80
- "normalized": false,
81
- "rstrip": false,
82
- "single_word": false,
83
- "special": true
84
- },
85
- "151653": {
86
- "content": "<|vision_end|>",
87
- "lstrip": false,
88
- "normalized": false,
89
- "rstrip": false,
90
- "single_word": false,
91
- "special": true
92
- },
93
- "151654": {
94
- "content": "<|vision_pad|>",
95
- "lstrip": false,
96
- "normalized": false,
97
- "rstrip": false,
98
- "single_word": false,
99
- "special": true
100
- },
101
- "151655": {
102
- "content": "<|image_pad|>",
103
- "lstrip": false,
104
- "normalized": false,
105
- "rstrip": false,
106
- "single_word": false,
107
- "special": true
108
- },
109
- "151656": {
110
- "content": "<|video_pad|>",
111
- "lstrip": false,
112
- "normalized": false,
113
- "rstrip": false,
114
- "single_word": false,
115
- "special": true
116
- },
117
- "151657": {
118
- "content": "<tool_call>",
119
- "lstrip": false,
120
- "normalized": false,
121
- "rstrip": false,
122
- "single_word": false,
123
- "special": false
124
- },
125
- "151658": {
126
- "content": "</tool_call>",
127
- "lstrip": false,
128
- "normalized": false,
129
- "rstrip": false,
130
- "single_word": false,
131
- "special": false
132
- },
133
- "151659": {
134
- "content": "<|fim_prefix|>",
135
- "lstrip": false,
136
- "normalized": false,
137
- "rstrip": false,
138
- "single_word": false,
139
- "special": false
140
- },
141
- "151660": {
142
- "content": "<|fim_middle|>",
143
- "lstrip": false,
144
- "normalized": false,
145
- "rstrip": false,
146
- "single_word": false,
147
- "special": false
148
- },
149
- "151661": {
150
- "content": "<|fim_suffix|>",
151
- "lstrip": false,
152
- "normalized": false,
153
- "rstrip": false,
154
- "single_word": false,
155
- "special": false
156
- },
157
- "151662": {
158
- "content": "<|fim_pad|>",
159
- "lstrip": false,
160
- "normalized": false,
161
- "rstrip": false,
162
- "single_word": false,
163
- "special": false
164
- },
165
- "151663": {
166
- "content": "<|repo_name|>",
167
- "lstrip": false,
168
- "normalized": false,
169
- "rstrip": false,
170
- "single_word": false,
171
- "special": false
172
- },
173
- "151664": {
174
- "content": "<|file_sep|>",
175
- "lstrip": false,
176
- "normalized": false,
177
- "rstrip": false,
178
- "single_word": false,
179
- "special": false
180
- },
181
- "151665": {
182
- "content": "<tool_response>",
183
- "lstrip": false,
184
- "normalized": false,
185
- "rstrip": false,
186
- "single_word": false,
187
- "special": false
188
- },
189
- "151666": {
190
- "content": "</tool_response>",
191
- "lstrip": false,
192
- "normalized": false,
193
- "rstrip": false,
194
- "single_word": false,
195
- "special": false
196
- },
197
- "151667": {
198
- "content": "<think>",
199
- "lstrip": false,
200
- "normalized": false,
201
- "rstrip": false,
202
- "single_word": false,
203
- "special": false
204
- },
205
- "151668": {
206
- "content": "</think>",
207
- "lstrip": false,
208
- "normalized": false,
209
- "rstrip": false,
210
- "single_word": false,
211
- "special": false
212
- }
213
- },
214
- "additional_special_tokens": [
215
- "<|im_start|>",
216
- "<|im_end|>",
217
- "<|object_ref_start|>",
218
- "<|object_ref_end|>",
219
- "<|box_start|>",
220
- "<|box_end|>",
221
- "<|quad_start|>",
222
- "<|quad_end|>",
223
- "<|vision_start|>",
224
- "<|vision_end|>",
225
- "<|vision_pad|>",
226
- "<|image_pad|>",
227
- "<|video_pad|>"
228
- ],
229
- "bos_token": null,
230
- "clean_up_tokenization_spaces": false,
231
- "eos_token": "<|im_end|>",
232
- "errors": "replace",
233
- "extra_special_tokens": {},
234
- "model_max_length": 131072,
235
- "pad_token": "<|endoftext|>",
236
- "split_special_tokens": false,
237
- "tokenizer_class": "Qwen2Tokenizer",
238
- "unk_token": null
239
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoints/trainer_state.json DELETED
@@ -1,2239 +0,0 @@
1
- {
2
- "best_global_step": null,
3
- "best_metric": null,
4
- "best_model_checkpoint": null,
5
- "epoch": 10.0,
6
- "eval_steps": 500,
7
- "global_step": 3140,
8
- "is_hyper_param_search": false,
9
- "is_local_process_zero": true,
10
- "is_world_process_zero": true,
11
- "log_history": [
12
- {
13
- "epoch": 0.003189792663476874,
14
- "grad_norm": 1.5252034664154053,
15
- "learning_rate": 0.0,
16
- "loss": 5.7289,
17
- "step": 1
18
- },
19
- {
20
- "epoch": 0.03189792663476874,
21
- "grad_norm": 1.3174983263015747,
22
- "learning_rate": 3.6e-05,
23
- "loss": 5.6677,
24
- "step": 10
25
- },
26
- {
27
- "epoch": 0.06379585326953748,
28
- "grad_norm": 0.8419873714447021,
29
- "learning_rate": 7.6e-05,
30
- "loss": 5.3345,
31
- "step": 20
32
- },
33
- {
34
- "epoch": 0.09569377990430622,
35
- "grad_norm": 0.501588761806488,
36
- "learning_rate": 0.000116,
37
- "loss": 4.9165,
38
- "step": 30
39
- },
40
- {
41
- "epoch": 0.12759170653907495,
42
- "grad_norm": 0.6858522891998291,
43
- "learning_rate": 0.00015600000000000002,
44
- "loss": 4.6209,
45
- "step": 40
46
- },
47
- {
48
- "epoch": 0.1594896331738437,
49
- "grad_norm": 0.8215904235839844,
50
- "learning_rate": 0.000196,
51
- "loss": 4.4453,
52
- "step": 50
53
- },
54
- {
55
- "epoch": 0.19138755980861244,
56
- "grad_norm": 1.1306647062301636,
57
- "learning_rate": 0.00019941747572815535,
58
- "loss": 4.3048,
59
- "step": 60
60
- },
61
- {
62
- "epoch": 0.22328548644338117,
63
- "grad_norm": 0.7509264945983887,
64
- "learning_rate": 0.00019877022653721685,
65
- "loss": 4.2047,
66
- "step": 70
67
- },
68
- {
69
- "epoch": 0.2551834130781499,
70
- "grad_norm": 0.7645938992500305,
71
- "learning_rate": 0.00019812297734627833,
72
- "loss": 4.1461,
73
- "step": 80
74
- },
75
- {
76
- "epoch": 0.28708133971291866,
77
- "grad_norm": 0.7448317408561707,
78
- "learning_rate": 0.0001974757281553398,
79
- "loss": 4.1107,
80
- "step": 90
81
- },
82
- {
83
- "epoch": 0.3189792663476874,
84
- "grad_norm": 1.1897181272506714,
85
- "learning_rate": 0.0001968284789644013,
86
- "loss": 4.0805,
87
- "step": 100
88
- },
89
- {
90
- "epoch": 0.3508771929824561,
91
- "grad_norm": 0.9169389605522156,
92
- "learning_rate": 0.0001961812297734628,
93
- "loss": 4.0398,
94
- "step": 110
95
- },
96
- {
97
- "epoch": 0.3827751196172249,
98
- "grad_norm": 1.102384090423584,
99
- "learning_rate": 0.0001955339805825243,
100
- "loss": 4.0247,
101
- "step": 120
102
- },
103
- {
104
- "epoch": 0.41467304625199364,
105
- "grad_norm": 0.8408002257347107,
106
- "learning_rate": 0.00019488673139158577,
107
- "loss": 4.0037,
108
- "step": 130
109
- },
110
- {
111
- "epoch": 0.44657097288676234,
112
- "grad_norm": 0.8749285936355591,
113
- "learning_rate": 0.00019423948220064724,
114
- "loss": 3.9804,
115
- "step": 140
116
- },
117
- {
118
- "epoch": 0.4784688995215311,
119
- "grad_norm": 1.2045485973358154,
120
- "learning_rate": 0.00019359223300970875,
121
- "loss": 3.9795,
122
- "step": 150
123
- },
124
- {
125
- "epoch": 0.5103668261562998,
126
- "grad_norm": 1.443690538406372,
127
- "learning_rate": 0.00019294498381877025,
128
- "loss": 3.9416,
129
- "step": 160
130
- },
131
- {
132
- "epoch": 0.5422647527910686,
133
- "grad_norm": 1.188532829284668,
134
- "learning_rate": 0.0001922977346278317,
135
- "loss": 3.9073,
136
- "step": 170
137
- },
138
- {
139
- "epoch": 0.5741626794258373,
140
- "grad_norm": 1.053863525390625,
141
- "learning_rate": 0.0001916504854368932,
142
- "loss": 3.888,
143
- "step": 180
144
- },
145
- {
146
- "epoch": 0.6060606060606061,
147
- "grad_norm": 1.6133322715759277,
148
- "learning_rate": 0.0001910032362459547,
149
- "loss": 3.8654,
150
- "step": 190
151
- },
152
- {
153
- "epoch": 0.6379585326953748,
154
- "grad_norm": 1.6843335628509521,
155
- "learning_rate": 0.0001903559870550162,
156
- "loss": 3.8268,
157
- "step": 200
158
- },
159
- {
160
- "epoch": 0.6698564593301436,
161
- "grad_norm": 2.04089617729187,
162
- "learning_rate": 0.00018970873786407767,
163
- "loss": 3.8068,
164
- "step": 210
165
- },
166
- {
167
- "epoch": 0.7017543859649122,
168
- "grad_norm": 1.8679882287979126,
169
- "learning_rate": 0.00018906148867313917,
170
- "loss": 3.7658,
171
- "step": 220
172
- },
173
- {
174
- "epoch": 0.733652312599681,
175
- "grad_norm": 1.8345612287521362,
176
- "learning_rate": 0.00018841423948220065,
177
- "loss": 3.7555,
178
- "step": 230
179
- },
180
- {
181
- "epoch": 0.7655502392344498,
182
- "grad_norm": 1.8040602207183838,
183
- "learning_rate": 0.00018776699029126215,
184
- "loss": 3.7112,
185
- "step": 240
186
- },
187
- {
188
- "epoch": 0.7974481658692185,
189
- "grad_norm": 2.027642250061035,
190
- "learning_rate": 0.00018711974110032366,
191
- "loss": 3.7088,
192
- "step": 250
193
- },
194
- {
195
- "epoch": 0.8293460925039873,
196
- "grad_norm": 2.0571959018707275,
197
- "learning_rate": 0.0001864724919093851,
198
- "loss": 3.6897,
199
- "step": 260
200
- },
201
- {
202
- "epoch": 0.861244019138756,
203
- "grad_norm": 2.0399420261383057,
204
- "learning_rate": 0.0001858252427184466,
205
- "loss": 3.6013,
206
- "step": 270
207
- },
208
- {
209
- "epoch": 0.8931419457735247,
210
- "grad_norm": 1.7937105894088745,
211
- "learning_rate": 0.0001851779935275081,
212
- "loss": 3.5909,
213
- "step": 280
214
- },
215
- {
216
- "epoch": 0.9250398724082934,
217
- "grad_norm": 2.259948968887329,
218
- "learning_rate": 0.0001845307443365696,
219
- "loss": 3.5773,
220
- "step": 290
221
- },
222
- {
223
- "epoch": 0.9569377990430622,
224
- "grad_norm": 3.999640941619873,
225
- "learning_rate": 0.00018388349514563107,
226
- "loss": 3.5398,
227
- "step": 300
228
- },
229
- {
230
- "epoch": 0.988835725677831,
231
- "grad_norm": 2.760211706161499,
232
- "learning_rate": 0.00018323624595469257,
233
- "loss": 3.4636,
234
- "step": 310
235
- },
236
- {
237
- "epoch": 1.0191387559808613,
238
- "grad_norm": 3.569462537765503,
239
- "learning_rate": 0.00018258899676375405,
240
- "loss": 3.208,
241
- "step": 320
242
- },
243
- {
244
- "epoch": 1.0510366826156299,
245
- "grad_norm": 5.000574111938477,
246
- "learning_rate": 0.00018194174757281555,
247
- "loss": 3.3652,
248
- "step": 330
249
- },
250
- {
251
- "epoch": 1.0829346092503986,
252
- "grad_norm": 3.697556734085083,
253
- "learning_rate": 0.00018129449838187703,
254
- "loss": 3.2778,
255
- "step": 340
256
- },
257
- {
258
- "epoch": 1.1148325358851674,
259
- "grad_norm": 3.266451835632324,
260
- "learning_rate": 0.0001806472491909385,
261
- "loss": 3.1285,
262
- "step": 350
263
- },
264
- {
265
- "epoch": 1.1467304625199362,
266
- "grad_norm": 3.4669878482818604,
267
- "learning_rate": 0.00018,
268
- "loss": 3.1121,
269
- "step": 360
270
- },
271
- {
272
- "epoch": 1.178628389154705,
273
- "grad_norm": 3.9828436374664307,
274
- "learning_rate": 0.0001793527508090615,
275
- "loss": 3.0454,
276
- "step": 370
277
- },
278
- {
279
- "epoch": 1.2105263157894737,
280
- "grad_norm": 4.253013610839844,
281
- "learning_rate": 0.000178705501618123,
282
- "loss": 2.9481,
283
- "step": 380
284
- },
285
- {
286
- "epoch": 1.2424242424242424,
287
- "grad_norm": 3.906336784362793,
288
- "learning_rate": 0.00017805825242718447,
289
- "loss": 2.9171,
290
- "step": 390
291
- },
292
- {
293
- "epoch": 1.2743221690590112,
294
- "grad_norm": 5.212265968322754,
295
- "learning_rate": 0.00017741100323624595,
296
- "loss": 2.7956,
297
- "step": 400
298
- },
299
- {
300
- "epoch": 1.30622009569378,
301
- "grad_norm": 5.57565975189209,
302
- "learning_rate": 0.00017676375404530745,
303
- "loss": 2.7399,
304
- "step": 410
305
- },
306
- {
307
- "epoch": 1.3381180223285487,
308
- "grad_norm": 7.32366418838501,
309
- "learning_rate": 0.00017611650485436896,
310
- "loss": 2.5755,
311
- "step": 420
312
- },
313
- {
314
- "epoch": 1.3700159489633175,
315
- "grad_norm": 5.175110816955566,
316
- "learning_rate": 0.00017546925566343043,
317
- "loss": 2.44,
318
- "step": 430
319
- },
320
- {
321
- "epoch": 1.401913875598086,
322
- "grad_norm": 6.860726356506348,
323
- "learning_rate": 0.0001748220064724919,
324
- "loss": 2.4152,
325
- "step": 440
326
- },
327
- {
328
- "epoch": 1.4338118022328548,
329
- "grad_norm": 6.029066562652588,
330
- "learning_rate": 0.0001741747572815534,
331
- "loss": 2.3313,
332
- "step": 450
333
- },
334
- {
335
- "epoch": 1.4657097288676235,
336
- "grad_norm": 5.50711727142334,
337
- "learning_rate": 0.0001735275080906149,
338
- "loss": 2.2367,
339
- "step": 460
340
- },
341
- {
342
- "epoch": 1.4976076555023923,
343
- "grad_norm": 7.61848258972168,
344
- "learning_rate": 0.0001728802588996764,
345
- "loss": 2.1664,
346
- "step": 470
347
- },
348
- {
349
- "epoch": 1.529505582137161,
350
- "grad_norm": 8.040018081665039,
351
- "learning_rate": 0.00017223300970873787,
352
- "loss": 1.9997,
353
- "step": 480
354
- },
355
- {
356
- "epoch": 1.5614035087719298,
357
- "grad_norm": 7.164926528930664,
358
- "learning_rate": 0.00017158576051779935,
359
- "loss": 1.9375,
360
- "step": 490
361
- },
362
- {
363
- "epoch": 1.5933014354066986,
364
- "grad_norm": 8.465059280395508,
365
- "learning_rate": 0.00017093851132686085,
366
- "loss": 1.7925,
367
- "step": 500
368
- },
369
- {
370
- "epoch": 1.6251993620414673,
371
- "grad_norm": 9.419471740722656,
372
- "learning_rate": 0.00017029126213592236,
373
- "loss": 1.8445,
374
- "step": 510
375
- },
376
- {
377
- "epoch": 1.657097288676236,
378
- "grad_norm": 8.507338523864746,
379
- "learning_rate": 0.00016964401294498383,
380
- "loss": 1.7681,
381
- "step": 520
382
- },
383
- {
384
- "epoch": 1.6889952153110048,
385
- "grad_norm": 9.474655151367188,
386
- "learning_rate": 0.0001689967637540453,
387
- "loss": 1.5889,
388
- "step": 530
389
- },
390
- {
391
- "epoch": 1.7208931419457736,
392
- "grad_norm": 9.287930488586426,
393
- "learning_rate": 0.00016834951456310682,
394
- "loss": 1.5392,
395
- "step": 540
396
- },
397
- {
398
- "epoch": 1.7527910685805423,
399
- "grad_norm": 6.526854991912842,
400
- "learning_rate": 0.0001677022653721683,
401
- "loss": 1.4232,
402
- "step": 550
403
- },
404
- {
405
- "epoch": 1.784688995215311,
406
- "grad_norm": 8.358418464660645,
407
- "learning_rate": 0.0001670550161812298,
408
- "loss": 1.2953,
409
- "step": 560
410
- },
411
- {
412
- "epoch": 1.8165869218500799,
413
- "grad_norm": 7.086184978485107,
414
- "learning_rate": 0.00016640776699029127,
415
- "loss": 1.3161,
416
- "step": 570
417
- },
418
- {
419
- "epoch": 1.8484848484848486,
420
- "grad_norm": 8.247760772705078,
421
- "learning_rate": 0.00016576051779935275,
422
- "loss": 1.0697,
423
- "step": 580
424
- },
425
- {
426
- "epoch": 1.8803827751196174,
427
- "grad_norm": 7.4881720542907715,
428
- "learning_rate": 0.00016511326860841425,
429
- "loss": 1.0045,
430
- "step": 590
431
- },
432
- {
433
- "epoch": 1.912280701754386,
434
- "grad_norm": 9.438504219055176,
435
- "learning_rate": 0.00016446601941747573,
436
- "loss": 0.9993,
437
- "step": 600
438
- },
439
- {
440
- "epoch": 1.9441786283891547,
441
- "grad_norm": 7.289414882659912,
442
- "learning_rate": 0.0001638187702265372,
443
- "loss": 0.8662,
444
- "step": 610
445
- },
446
- {
447
- "epoch": 1.9760765550239234,
448
- "grad_norm": 9.634711265563965,
449
- "learning_rate": 0.0001631715210355987,
450
- "loss": 0.9138,
451
- "step": 620
452
- },
453
- {
454
- "epoch": 2.006379585326954,
455
- "grad_norm": 7.288631916046143,
456
- "learning_rate": 0.0001625242718446602,
457
- "loss": 0.8087,
458
- "step": 630
459
- },
460
- {
461
- "epoch": 2.0382775119617227,
462
- "grad_norm": 11.611783027648926,
463
- "learning_rate": 0.0001618770226537217,
464
- "loss": 0.6835,
465
- "step": 640
466
- },
467
- {
468
- "epoch": 2.0701754385964914,
469
- "grad_norm": 7.48546028137207,
470
- "learning_rate": 0.00016122977346278317,
471
- "loss": 0.5253,
472
- "step": 650
473
- },
474
- {
475
- "epoch": 2.1020733652312598,
476
- "grad_norm": 8.046589851379395,
477
- "learning_rate": 0.00016058252427184465,
478
- "loss": 0.555,
479
- "step": 660
480
- },
481
- {
482
- "epoch": 2.1339712918660285,
483
- "grad_norm": 5.977556228637695,
484
- "learning_rate": 0.00015993527508090615,
485
- "loss": 0.5774,
486
- "step": 670
487
- },
488
- {
489
- "epoch": 2.1658692185007973,
490
- "grad_norm": 7.377207279205322,
491
- "learning_rate": 0.00015928802588996766,
492
- "loss": 0.5707,
493
- "step": 680
494
- },
495
- {
496
- "epoch": 2.197767145135566,
497
- "grad_norm": 6.869225025177002,
498
- "learning_rate": 0.00015864077669902913,
499
- "loss": 0.5913,
500
- "step": 690
501
- },
502
- {
503
- "epoch": 2.229665071770335,
504
- "grad_norm": 5.870683193206787,
505
- "learning_rate": 0.0001579935275080906,
506
- "loss": 0.4496,
507
- "step": 700
508
- },
509
- {
510
- "epoch": 2.2615629984051036,
511
- "grad_norm": 7.628873348236084,
512
- "learning_rate": 0.00015734627831715212,
513
- "loss": 0.4418,
514
- "step": 710
515
- },
516
- {
517
- "epoch": 2.2934609250398723,
518
- "grad_norm": 7.419798374176025,
519
- "learning_rate": 0.0001566990291262136,
520
- "loss": 0.3939,
521
- "step": 720
522
- },
523
- {
524
- "epoch": 2.325358851674641,
525
- "grad_norm": 5.196732521057129,
526
- "learning_rate": 0.0001560517799352751,
527
- "loss": 0.3817,
528
- "step": 730
529
- },
530
- {
531
- "epoch": 2.35725677830941,
532
- "grad_norm": 7.466589450836182,
533
- "learning_rate": 0.00015540453074433657,
534
- "loss": 0.3564,
535
- "step": 740
536
- },
537
- {
538
- "epoch": 2.3891547049441786,
539
- "grad_norm": 4.85908317565918,
540
- "learning_rate": 0.00015475728155339805,
541
- "loss": 0.3566,
542
- "step": 750
543
- },
544
- {
545
- "epoch": 2.4210526315789473,
546
- "grad_norm": 6.417631149291992,
547
- "learning_rate": 0.00015411003236245955,
548
- "loss": 0.3637,
549
- "step": 760
550
- },
551
- {
552
- "epoch": 2.452950558213716,
553
- "grad_norm": 8.523224830627441,
554
- "learning_rate": 0.00015346278317152106,
555
- "loss": 0.453,
556
- "step": 770
557
- },
558
- {
559
- "epoch": 2.484848484848485,
560
- "grad_norm": 5.428617477416992,
561
- "learning_rate": 0.00015281553398058254,
562
- "loss": 0.3353,
563
- "step": 780
564
- },
565
- {
566
- "epoch": 2.5167464114832536,
567
- "grad_norm": 4.765276908874512,
568
- "learning_rate": 0.000152168284789644,
569
- "loss": 0.3291,
570
- "step": 790
571
- },
572
- {
573
- "epoch": 2.5486443381180224,
574
- "grad_norm": 6.175685405731201,
575
- "learning_rate": 0.00015152103559870552,
576
- "loss": 0.2628,
577
- "step": 800
578
- },
579
- {
580
- "epoch": 2.580542264752791,
581
- "grad_norm": 7.138719081878662,
582
- "learning_rate": 0.000150873786407767,
583
- "loss": 0.2652,
584
- "step": 810
585
- },
586
- {
587
- "epoch": 2.61244019138756,
588
- "grad_norm": 5.668585777282715,
589
- "learning_rate": 0.0001502265372168285,
590
- "loss": 0.252,
591
- "step": 820
592
- },
593
- {
594
- "epoch": 2.6443381180223287,
595
- "grad_norm": 5.076392650604248,
596
- "learning_rate": 0.00014957928802588998,
597
- "loss": 0.2438,
598
- "step": 830
599
- },
600
- {
601
- "epoch": 2.6762360446570974,
602
- "grad_norm": 8.504268646240234,
603
- "learning_rate": 0.00014893203883495145,
604
- "loss": 0.3538,
605
- "step": 840
606
- },
607
- {
608
- "epoch": 2.708133971291866,
609
- "grad_norm": 4.550518989562988,
610
- "learning_rate": 0.00014828478964401296,
611
- "loss": 0.2585,
612
- "step": 850
613
- },
614
- {
615
- "epoch": 2.740031897926635,
616
- "grad_norm": 6.09181547164917,
617
- "learning_rate": 0.00014763754045307443,
618
- "loss": 0.2586,
619
- "step": 860
620
- },
621
- {
622
- "epoch": 2.7719298245614032,
623
- "grad_norm": 7.383531093597412,
624
- "learning_rate": 0.00014699029126213594,
625
- "loss": 0.2497,
626
- "step": 870
627
- },
628
- {
629
- "epoch": 2.803827751196172,
630
- "grad_norm": 5.605717182159424,
631
- "learning_rate": 0.00014634304207119741,
632
- "loss": 0.2285,
633
- "step": 880
634
- },
635
- {
636
- "epoch": 2.8357256778309408,
637
- "grad_norm": 7.412971496582031,
638
- "learning_rate": 0.0001456957928802589,
639
- "loss": 0.257,
640
- "step": 890
641
- },
642
- {
643
- "epoch": 2.8676236044657095,
644
- "grad_norm": 6.855090141296387,
645
- "learning_rate": 0.0001450485436893204,
646
- "loss": 0.2347,
647
- "step": 900
648
- },
649
- {
650
- "epoch": 2.8995215311004783,
651
- "grad_norm": 4.776122570037842,
652
- "learning_rate": 0.0001444012944983819,
653
- "loss": 0.2527,
654
- "step": 910
655
- },
656
- {
657
- "epoch": 2.931419457735247,
658
- "grad_norm": 3.841705560684204,
659
- "learning_rate": 0.00014375404530744335,
660
- "loss": 0.192,
661
- "step": 920
662
- },
663
- {
664
- "epoch": 2.963317384370016,
665
- "grad_norm": 4.253203392028809,
666
- "learning_rate": 0.00014310679611650485,
667
- "loss": 0.2104,
668
- "step": 930
669
- },
670
- {
671
- "epoch": 2.9952153110047846,
672
- "grad_norm": 5.285226345062256,
673
- "learning_rate": 0.00014245954692556636,
674
- "loss": 0.1655,
675
- "step": 940
676
- },
677
- {
678
- "epoch": 3.025518341307815,
679
- "grad_norm": 3.939434051513672,
680
- "learning_rate": 0.00014181229773462784,
681
- "loss": 0.1744,
682
- "step": 950
683
- },
684
- {
685
- "epoch": 3.0574162679425836,
686
- "grad_norm": 4.514401435852051,
687
- "learning_rate": 0.00014116504854368934,
688
- "loss": 0.179,
689
- "step": 960
690
- },
691
- {
692
- "epoch": 3.0893141945773523,
693
- "grad_norm": 6.335837364196777,
694
- "learning_rate": 0.00014051779935275082,
695
- "loss": 0.2103,
696
- "step": 970
697
- },
698
- {
699
- "epoch": 3.121212121212121,
700
- "grad_norm": 3.401519775390625,
701
- "learning_rate": 0.0001398705501618123,
702
- "loss": 0.1717,
703
- "step": 980
704
- },
705
- {
706
- "epoch": 3.15311004784689,
707
- "grad_norm": 3.501477003097534,
708
- "learning_rate": 0.0001392233009708738,
709
- "loss": 0.149,
710
- "step": 990
711
- },
712
- {
713
- "epoch": 3.1850079744816586,
714
- "grad_norm": 4.168037414550781,
715
- "learning_rate": 0.0001385760517799353,
716
- "loss": 0.1557,
717
- "step": 1000
718
- },
719
- {
720
- "epoch": 3.2169059011164274,
721
- "grad_norm": 5.530002117156982,
722
- "learning_rate": 0.00013792880258899675,
723
- "loss": 0.1878,
724
- "step": 1010
725
- },
726
- {
727
- "epoch": 3.248803827751196,
728
- "grad_norm": 6.048915863037109,
729
- "learning_rate": 0.00013728155339805826,
730
- "loss": 0.1915,
731
- "step": 1020
732
- },
733
- {
734
- "epoch": 3.280701754385965,
735
- "grad_norm": 3.8435139656066895,
736
- "learning_rate": 0.00013663430420711976,
737
- "loss": 0.1429,
738
- "step": 1030
739
- },
740
- {
741
- "epoch": 3.3125996810207337,
742
- "grad_norm": 5.1407952308654785,
743
- "learning_rate": 0.00013598705501618124,
744
- "loss": 0.1486,
745
- "step": 1040
746
- },
747
- {
748
- "epoch": 3.3444976076555024,
749
- "grad_norm": 3.7761664390563965,
750
- "learning_rate": 0.00013533980582524271,
751
- "loss": 0.1758,
752
- "step": 1050
753
- },
754
- {
755
- "epoch": 3.376395534290271,
756
- "grad_norm": 4.0396952629089355,
757
- "learning_rate": 0.00013469255663430422,
758
- "loss": 0.1692,
759
- "step": 1060
760
- },
761
- {
762
- "epoch": 3.40829346092504,
763
- "grad_norm": 6.123923301696777,
764
- "learning_rate": 0.0001340453074433657,
765
- "loss": 0.1761,
766
- "step": 1070
767
- },
768
- {
769
- "epoch": 3.4401913875598087,
770
- "grad_norm": 3.7022392749786377,
771
- "learning_rate": 0.0001333980582524272,
772
- "loss": 0.1338,
773
- "step": 1080
774
- },
775
- {
776
- "epoch": 3.4720893141945774,
777
- "grad_norm": 3.5536398887634277,
778
- "learning_rate": 0.00013275080906148868,
779
- "loss": 0.1375,
780
- "step": 1090
781
- },
782
- {
783
- "epoch": 3.503987240829346,
784
- "grad_norm": 5.543455123901367,
785
- "learning_rate": 0.00013210355987055015,
786
- "loss": 0.1477,
787
- "step": 1100
788
- },
789
- {
790
- "epoch": 3.535885167464115,
791
- "grad_norm": 3.780407667160034,
792
- "learning_rate": 0.00013145631067961166,
793
- "loss": 0.168,
794
- "step": 1110
795
- },
796
- {
797
- "epoch": 3.5677830940988837,
798
- "grad_norm": 3.6479198932647705,
799
- "learning_rate": 0.00013080906148867314,
800
- "loss": 0.115,
801
- "step": 1120
802
- },
803
- {
804
- "epoch": 3.5996810207336525,
805
- "grad_norm": 4.027154445648193,
806
- "learning_rate": 0.00013016181229773464,
807
- "loss": 0.1251,
808
- "step": 1130
809
- },
810
- {
811
- "epoch": 3.6315789473684212,
812
- "grad_norm": 3.59836745262146,
813
- "learning_rate": 0.00012951456310679612,
814
- "loss": 0.1294,
815
- "step": 1140
816
- },
817
- {
818
- "epoch": 3.66347687400319,
819
- "grad_norm": 3.4946069717407227,
820
- "learning_rate": 0.0001288673139158576,
821
- "loss": 0.1161,
822
- "step": 1150
823
- },
824
- {
825
- "epoch": 3.6953748006379588,
826
- "grad_norm": 3.2438859939575195,
827
- "learning_rate": 0.0001282200647249191,
828
- "loss": 0.1003,
829
- "step": 1160
830
- },
831
- {
832
- "epoch": 3.7272727272727275,
833
- "grad_norm": 3.518580675125122,
834
- "learning_rate": 0.0001275728155339806,
835
- "loss": 0.1088,
836
- "step": 1170
837
- },
838
- {
839
- "epoch": 3.7591706539074963,
840
- "grad_norm": 3.3931541442871094,
841
- "learning_rate": 0.00012692556634304208,
842
- "loss": 0.1162,
843
- "step": 1180
844
- },
845
- {
846
- "epoch": 3.7910685805422646,
847
- "grad_norm": 4.836375713348389,
848
- "learning_rate": 0.00012627831715210356,
849
- "loss": 0.1206,
850
- "step": 1190
851
- },
852
- {
853
- "epoch": 3.8229665071770333,
854
- "grad_norm": 3.4532904624938965,
855
- "learning_rate": 0.00012563106796116506,
856
- "loss": 0.0998,
857
- "step": 1200
858
- },
859
- {
860
- "epoch": 3.854864433811802,
861
- "grad_norm": 3.0089080333709717,
862
- "learning_rate": 0.00012498381877022654,
863
- "loss": 0.1325,
864
- "step": 1210
865
- },
866
- {
867
- "epoch": 3.886762360446571,
868
- "grad_norm": 2.9546520709991455,
869
- "learning_rate": 0.00012433656957928804,
870
- "loss": 0.1073,
871
- "step": 1220
872
- },
873
- {
874
- "epoch": 3.9186602870813396,
875
- "grad_norm": 4.0866851806640625,
876
- "learning_rate": 0.00012368932038834952,
877
- "loss": 0.1135,
878
- "step": 1230
879
- },
880
- {
881
- "epoch": 3.9505582137161084,
882
- "grad_norm": 2.4372901916503906,
883
- "learning_rate": 0.000123042071197411,
884
- "loss": 0.1078,
885
- "step": 1240
886
- },
887
- {
888
- "epoch": 3.982456140350877,
889
- "grad_norm": 3.0094428062438965,
890
- "learning_rate": 0.0001223948220064725,
891
- "loss": 0.0926,
892
- "step": 1250
893
- },
894
- {
895
- "epoch": 4.012759170653908,
896
- "grad_norm": 3.015343189239502,
897
- "learning_rate": 0.00012174757281553399,
898
- "loss": 0.0812,
899
- "step": 1260
900
- },
901
- {
902
- "epoch": 4.044657097288677,
903
- "grad_norm": 3.3010330200195312,
904
- "learning_rate": 0.00012110032362459547,
905
- "loss": 0.0912,
906
- "step": 1270
907
- },
908
- {
909
- "epoch": 4.076555023923445,
910
- "grad_norm": 3.2524020671844482,
911
- "learning_rate": 0.00012045307443365696,
912
- "loss": 0.0919,
913
- "step": 1280
914
- },
915
- {
916
- "epoch": 4.108452950558214,
917
- "grad_norm": 3.144233465194702,
918
- "learning_rate": 0.00011980582524271846,
919
- "loss": 0.0984,
920
- "step": 1290
921
- },
922
- {
923
- "epoch": 4.140350877192983,
924
- "grad_norm": 3.6197972297668457,
925
- "learning_rate": 0.00011915857605177994,
926
- "loss": 0.0888,
927
- "step": 1300
928
- },
929
- {
930
- "epoch": 4.172248803827751,
931
- "grad_norm": 3.464054584503174,
932
- "learning_rate": 0.00011851132686084143,
933
- "loss": 0.0948,
934
- "step": 1310
935
- },
936
- {
937
- "epoch": 4.2041467304625195,
938
- "grad_norm": 3.589921474456787,
939
- "learning_rate": 0.00011786407766990293,
940
- "loss": 0.1008,
941
- "step": 1320
942
- },
943
- {
944
- "epoch": 4.236044657097288,
945
- "grad_norm": 2.547912120819092,
946
- "learning_rate": 0.0001172168284789644,
947
- "loss": 0.0865,
948
- "step": 1330
949
- },
950
- {
951
- "epoch": 4.267942583732057,
952
- "grad_norm": 2.76481032371521,
953
- "learning_rate": 0.0001165695792880259,
954
- "loss": 0.0879,
955
- "step": 1340
956
- },
957
- {
958
- "epoch": 4.299840510366826,
959
- "grad_norm": 3.417572021484375,
960
- "learning_rate": 0.00011592233009708739,
961
- "loss": 0.0824,
962
- "step": 1350
963
- },
964
- {
965
- "epoch": 4.3317384370015946,
966
- "grad_norm": 2.9202117919921875,
967
- "learning_rate": 0.00011527508090614887,
968
- "loss": 0.0843,
969
- "step": 1360
970
- },
971
- {
972
- "epoch": 4.363636363636363,
973
- "grad_norm": 3.2221357822418213,
974
- "learning_rate": 0.00011462783171521036,
975
- "loss": 0.0935,
976
- "step": 1370
977
- },
978
- {
979
- "epoch": 4.395534290271132,
980
- "grad_norm": 3.54180645942688,
981
- "learning_rate": 0.00011398058252427184,
982
- "loss": 0.0932,
983
- "step": 1380
984
- },
985
- {
986
- "epoch": 4.427432216905901,
987
- "grad_norm": 2.438354253768921,
988
- "learning_rate": 0.00011333333333333334,
989
- "loss": 0.0787,
990
- "step": 1390
991
- },
992
- {
993
- "epoch": 4.45933014354067,
994
- "grad_norm": 1.9685388803482056,
995
- "learning_rate": 0.00011268608414239483,
996
- "loss": 0.0745,
997
- "step": 1400
998
- },
999
- {
1000
- "epoch": 4.491228070175438,
1001
- "grad_norm": 3.79945969581604,
1002
- "learning_rate": 0.00011203883495145631,
1003
- "loss": 0.0941,
1004
- "step": 1410
1005
- },
1006
- {
1007
- "epoch": 4.523125996810207,
1008
- "grad_norm": 2.6421890258789062,
1009
- "learning_rate": 0.0001113915857605178,
1010
- "loss": 0.0703,
1011
- "step": 1420
1012
- },
1013
- {
1014
- "epoch": 4.555023923444976,
1015
- "grad_norm": 2.041005849838257,
1016
- "learning_rate": 0.0001107443365695793,
1017
- "loss": 0.0739,
1018
- "step": 1430
1019
- },
1020
- {
1021
- "epoch": 4.586921850079745,
1022
- "grad_norm": 3.8904316425323486,
1023
- "learning_rate": 0.00011009708737864077,
1024
- "loss": 0.0698,
1025
- "step": 1440
1026
- },
1027
- {
1028
- "epoch": 4.618819776714513,
1029
- "grad_norm": 2.766702651977539,
1030
- "learning_rate": 0.00010944983818770227,
1031
- "loss": 0.0766,
1032
- "step": 1450
1033
- },
1034
- {
1035
- "epoch": 4.650717703349282,
1036
- "grad_norm": 2.6301941871643066,
1037
- "learning_rate": 0.00010880258899676376,
1038
- "loss": 0.0722,
1039
- "step": 1460
1040
- },
1041
- {
1042
- "epoch": 4.682615629984051,
1043
- "grad_norm": 3.7244162559509277,
1044
- "learning_rate": 0.00010815533980582524,
1045
- "loss": 0.0855,
1046
- "step": 1470
1047
- },
1048
- {
1049
- "epoch": 4.71451355661882,
1050
- "grad_norm": 2.076242446899414,
1051
- "learning_rate": 0.00010750809061488673,
1052
- "loss": 0.0625,
1053
- "step": 1480
1054
- },
1055
- {
1056
- "epoch": 4.746411483253588,
1057
- "grad_norm": 2.212062358856201,
1058
- "learning_rate": 0.00010686084142394823,
1059
- "loss": 0.0667,
1060
- "step": 1490
1061
- },
1062
- {
1063
- "epoch": 4.778309409888357,
1064
- "grad_norm": 3.0947024822235107,
1065
- "learning_rate": 0.00010621359223300971,
1066
- "loss": 0.0639,
1067
- "step": 1500
1068
- },
1069
- {
1070
- "epoch": 4.810207336523126,
1071
- "grad_norm": 3.310490369796753,
1072
- "learning_rate": 0.0001055663430420712,
1073
- "loss": 0.0821,
1074
- "step": 1510
1075
- },
1076
- {
1077
- "epoch": 4.842105263157895,
1078
- "grad_norm": 2.8293209075927734,
1079
- "learning_rate": 0.0001049190938511327,
1080
- "loss": 0.0733,
1081
- "step": 1520
1082
- },
1083
- {
1084
- "epoch": 4.8740031897926634,
1085
- "grad_norm": 2.975891590118408,
1086
- "learning_rate": 0.00010427184466019417,
1087
- "loss": 0.0686,
1088
- "step": 1530
1089
- },
1090
- {
1091
- "epoch": 4.905901116427432,
1092
- "grad_norm": 2.633657693862915,
1093
- "learning_rate": 0.00010362459546925567,
1094
- "loss": 0.0711,
1095
- "step": 1540
1096
- },
1097
- {
1098
- "epoch": 4.937799043062201,
1099
- "grad_norm": 2.947361469268799,
1100
- "learning_rate": 0.00010297734627831716,
1101
- "loss": 0.0686,
1102
- "step": 1550
1103
- },
1104
- {
1105
- "epoch": 4.96969696969697,
1106
- "grad_norm": 2.373194456100464,
1107
- "learning_rate": 0.00010233009708737864,
1108
- "loss": 0.0652,
1109
- "step": 1560
1110
- },
1111
- {
1112
- "epoch": 5.0,
1113
- "grad_norm": 1.3575212955474854,
1114
- "learning_rate": 0.00010168284789644013,
1115
- "loss": 0.0604,
1116
- "step": 1570
1117
- },
1118
- {
1119
- "epoch": 5.031897926634769,
1120
- "grad_norm": 2.9779140949249268,
1121
- "learning_rate": 0.00010103559870550164,
1122
- "loss": 0.0636,
1123
- "step": 1580
1124
- },
1125
- {
1126
- "epoch": 5.0637958532695375,
1127
- "grad_norm": 2.8484952449798584,
1128
- "learning_rate": 0.00010038834951456311,
1129
- "loss": 0.0558,
1130
- "step": 1590
1131
- },
1132
- {
1133
- "epoch": 5.095693779904306,
1134
- "grad_norm": 2.9866819381713867,
1135
- "learning_rate": 9.97411003236246e-05,
1136
- "loss": 0.0596,
1137
- "step": 1600
1138
- },
1139
- {
1140
- "epoch": 5.127591706539075,
1141
- "grad_norm": 2.2801356315612793,
1142
- "learning_rate": 9.90938511326861e-05,
1143
- "loss": 0.0557,
1144
- "step": 1610
1145
- },
1146
- {
1147
- "epoch": 5.159489633173844,
1148
- "grad_norm": 3.125821828842163,
1149
- "learning_rate": 9.844660194174757e-05,
1150
- "loss": 0.0671,
1151
- "step": 1620
1152
- },
1153
- {
1154
- "epoch": 5.1913875598086126,
1155
- "grad_norm": 2.4626238346099854,
1156
- "learning_rate": 9.779935275080908e-05,
1157
- "loss": 0.0615,
1158
- "step": 1630
1159
- },
1160
- {
1161
- "epoch": 5.223285486443381,
1162
- "grad_norm": 2.107201099395752,
1163
- "learning_rate": 9.715210355987055e-05,
1164
- "loss": 0.0597,
1165
- "step": 1640
1166
- },
1167
- {
1168
- "epoch": 5.25518341307815,
1169
- "grad_norm": 2.0662150382995605,
1170
- "learning_rate": 9.650485436893204e-05,
1171
- "loss": 0.0543,
1172
- "step": 1650
1173
- },
1174
- {
1175
- "epoch": 5.287081339712919,
1176
- "grad_norm": 2.728776454925537,
1177
- "learning_rate": 9.585760517799353e-05,
1178
- "loss": 0.0615,
1179
- "step": 1660
1180
- },
1181
- {
1182
- "epoch": 5.318979266347688,
1183
- "grad_norm": 2.2817022800445557,
1184
- "learning_rate": 9.521035598705502e-05,
1185
- "loss": 0.0513,
1186
- "step": 1670
1187
- },
1188
- {
1189
- "epoch": 5.350877192982456,
1190
- "grad_norm": 3.146657943725586,
1191
- "learning_rate": 9.45631067961165e-05,
1192
- "loss": 0.0523,
1193
- "step": 1680
1194
- },
1195
- {
1196
- "epoch": 5.382775119617225,
1197
- "grad_norm": 2.3867263793945312,
1198
- "learning_rate": 9.391585760517799e-05,
1199
- "loss": 0.0534,
1200
- "step": 1690
1201
- },
1202
- {
1203
- "epoch": 5.414673046251994,
1204
- "grad_norm": 2.2965664863586426,
1205
- "learning_rate": 9.326860841423948e-05,
1206
- "loss": 0.0495,
1207
- "step": 1700
1208
- },
1209
- {
1210
- "epoch": 5.446570972886763,
1211
- "grad_norm": 1.410491704940796,
1212
- "learning_rate": 9.262135922330097e-05,
1213
- "loss": 0.0523,
1214
- "step": 1710
1215
- },
1216
- {
1217
- "epoch": 5.478468899521531,
1218
- "grad_norm": 3.351827621459961,
1219
- "learning_rate": 9.197411003236246e-05,
1220
- "loss": 0.0516,
1221
- "step": 1720
1222
- },
1223
- {
1224
- "epoch": 5.5103668261563,
1225
- "grad_norm": 2.1586365699768066,
1226
- "learning_rate": 9.132686084142395e-05,
1227
- "loss": 0.0542,
1228
- "step": 1730
1229
- },
1230
- {
1231
- "epoch": 5.542264752791069,
1232
- "grad_norm": 1.751285433769226,
1233
- "learning_rate": 9.067961165048545e-05,
1234
- "loss": 0.0508,
1235
- "step": 1740
1236
- },
1237
- {
1238
- "epoch": 5.574162679425838,
1239
- "grad_norm": 1.4902950525283813,
1240
- "learning_rate": 9.003236245954692e-05,
1241
- "loss": 0.0547,
1242
- "step": 1750
1243
- },
1244
- {
1245
- "epoch": 5.606060606060606,
1246
- "grad_norm": 2.029918670654297,
1247
- "learning_rate": 8.938511326860843e-05,
1248
- "loss": 0.0561,
1249
- "step": 1760
1250
- },
1251
- {
1252
- "epoch": 5.637958532695375,
1253
- "grad_norm": 3.787614583969116,
1254
- "learning_rate": 8.87378640776699e-05,
1255
- "loss": 0.0528,
1256
- "step": 1770
1257
- },
1258
- {
1259
- "epoch": 5.669856459330144,
1260
- "grad_norm": 1.8037543296813965,
1261
- "learning_rate": 8.80906148867314e-05,
1262
- "loss": 0.0507,
1263
- "step": 1780
1264
- },
1265
- {
1266
- "epoch": 5.701754385964913,
1267
- "grad_norm": 2.3712103366851807,
1268
- "learning_rate": 8.744336569579288e-05,
1269
- "loss": 0.045,
1270
- "step": 1790
1271
- },
1272
- {
1273
- "epoch": 5.733652312599681,
1274
- "grad_norm": 1.5658912658691406,
1275
- "learning_rate": 8.679611650485438e-05,
1276
- "loss": 0.0472,
1277
- "step": 1800
1278
- },
1279
- {
1280
- "epoch": 5.76555023923445,
1281
- "grad_norm": 2.318931818008423,
1282
- "learning_rate": 8.614886731391587e-05,
1283
- "loss": 0.0484,
1284
- "step": 1810
1285
- },
1286
- {
1287
- "epoch": 5.797448165869218,
1288
- "grad_norm": 2.4893152713775635,
1289
- "learning_rate": 8.550161812297734e-05,
1290
- "loss": 0.0467,
1291
- "step": 1820
1292
- },
1293
- {
1294
- "epoch": 5.829346092503988,
1295
- "grad_norm": 1.9527428150177002,
1296
- "learning_rate": 8.485436893203885e-05,
1297
- "loss": 0.0498,
1298
- "step": 1830
1299
- },
1300
- {
1301
- "epoch": 5.861244019138756,
1302
- "grad_norm": 2.137164831161499,
1303
- "learning_rate": 8.420711974110032e-05,
1304
- "loss": 0.0446,
1305
- "step": 1840
1306
- },
1307
- {
1308
- "epoch": 5.893141945773524,
1309
- "grad_norm": 1.574310302734375,
1310
- "learning_rate": 8.355987055016181e-05,
1311
- "loss": 0.0471,
1312
- "step": 1850
1313
- },
1314
- {
1315
- "epoch": 5.925039872408293,
1316
- "grad_norm": 2.1740546226501465,
1317
- "learning_rate": 8.29126213592233e-05,
1318
- "loss": 0.0509,
1319
- "step": 1860
1320
- },
1321
- {
1322
- "epoch": 5.956937799043062,
1323
- "grad_norm": 2.2676613330841064,
1324
- "learning_rate": 8.22653721682848e-05,
1325
- "loss": 0.0473,
1326
- "step": 1870
1327
- },
1328
- {
1329
- "epoch": 5.988835725677831,
1330
- "grad_norm": 1.7788255214691162,
1331
- "learning_rate": 8.161812297734627e-05,
1332
- "loss": 0.0458,
1333
- "step": 1880
1334
- },
1335
- {
1336
- "epoch": 6.019138755980861,
1337
- "grad_norm": 2.5608322620391846,
1338
- "learning_rate": 8.097087378640778e-05,
1339
- "loss": 0.0412,
1340
- "step": 1890
1341
- },
1342
- {
1343
- "epoch": 6.05103668261563,
1344
- "grad_norm": 2.0553760528564453,
1345
- "learning_rate": 8.032362459546925e-05,
1346
- "loss": 0.0508,
1347
- "step": 1900
1348
- },
1349
- {
1350
- "epoch": 6.082934609250398,
1351
- "grad_norm": 1.3741275072097778,
1352
- "learning_rate": 7.967637540453075e-05,
1353
- "loss": 0.0382,
1354
- "step": 1910
1355
- },
1356
- {
1357
- "epoch": 6.114832535885167,
1358
- "grad_norm": 1.6682943105697632,
1359
- "learning_rate": 7.902912621359224e-05,
1360
- "loss": 0.0392,
1361
- "step": 1920
1362
- },
1363
- {
1364
- "epoch": 6.146730462519936,
1365
- "grad_norm": 1.4680795669555664,
1366
- "learning_rate": 7.838187702265373e-05,
1367
- "loss": 0.0431,
1368
- "step": 1930
1369
- },
1370
- {
1371
- "epoch": 6.178628389154705,
1372
- "grad_norm": 1.8964109420776367,
1373
- "learning_rate": 7.773462783171522e-05,
1374
- "loss": 0.0388,
1375
- "step": 1940
1376
- },
1377
- {
1378
- "epoch": 6.2105263157894735,
1379
- "grad_norm": 1.709599256515503,
1380
- "learning_rate": 7.70873786407767e-05,
1381
- "loss": 0.04,
1382
- "step": 1950
1383
- },
1384
- {
1385
- "epoch": 6.242424242424242,
1386
- "grad_norm": 1.761167287826538,
1387
- "learning_rate": 7.64401294498382e-05,
1388
- "loss": 0.0398,
1389
- "step": 1960
1390
- },
1391
- {
1392
- "epoch": 6.274322169059011,
1393
- "grad_norm": 2.034879207611084,
1394
- "learning_rate": 7.579288025889968e-05,
1395
- "loss": 0.0417,
1396
- "step": 1970
1397
- },
1398
- {
1399
- "epoch": 6.30622009569378,
1400
- "grad_norm": 1.5820090770721436,
1401
- "learning_rate": 7.514563106796117e-05,
1402
- "loss": 0.0411,
1403
- "step": 1980
1404
- },
1405
- {
1406
- "epoch": 6.3381180223285485,
1407
- "grad_norm": 1.5550072193145752,
1408
- "learning_rate": 7.449838187702266e-05,
1409
- "loss": 0.0365,
1410
- "step": 1990
1411
- },
1412
- {
1413
- "epoch": 6.370015948963317,
1414
- "grad_norm": 1.6271281242370605,
1415
- "learning_rate": 7.385113268608415e-05,
1416
- "loss": 0.0361,
1417
- "step": 2000
1418
- },
1419
- {
1420
- "epoch": 6.401913875598086,
1421
- "grad_norm": 1.7692902088165283,
1422
- "learning_rate": 7.320388349514564e-05,
1423
- "loss": 0.0364,
1424
- "step": 2010
1425
- },
1426
- {
1427
- "epoch": 6.433811802232855,
1428
- "grad_norm": 1.5776255130767822,
1429
- "learning_rate": 7.255663430420713e-05,
1430
- "loss": 0.0409,
1431
- "step": 2020
1432
- },
1433
- {
1434
- "epoch": 6.4657097288676235,
1435
- "grad_norm": 1.9021235704421997,
1436
- "learning_rate": 7.190938511326862e-05,
1437
- "loss": 0.0379,
1438
- "step": 2030
1439
- },
1440
- {
1441
- "epoch": 6.497607655502392,
1442
- "grad_norm": 1.342091679573059,
1443
- "learning_rate": 7.12621359223301e-05,
1444
- "loss": 0.0398,
1445
- "step": 2040
1446
- },
1447
- {
1448
- "epoch": 6.529505582137161,
1449
- "grad_norm": 2.0045480728149414,
1450
- "learning_rate": 7.06148867313916e-05,
1451
- "loss": 0.0403,
1452
- "step": 2050
1453
- },
1454
- {
1455
- "epoch": 6.56140350877193,
1456
- "grad_norm": 1.3946913480758667,
1457
- "learning_rate": 6.996763754045308e-05,
1458
- "loss": 0.0387,
1459
- "step": 2060
1460
- },
1461
- {
1462
- "epoch": 6.5933014354066986,
1463
- "grad_norm": 1.63865327835083,
1464
- "learning_rate": 6.932038834951457e-05,
1465
- "loss": 0.0377,
1466
- "step": 2070
1467
- },
1468
- {
1469
- "epoch": 6.625199362041467,
1470
- "grad_norm": 1.439233422279358,
1471
- "learning_rate": 6.867313915857604e-05,
1472
- "loss": 0.0376,
1473
- "step": 2080
1474
- },
1475
- {
1476
- "epoch": 6.657097288676236,
1477
- "grad_norm": 2.021103858947754,
1478
- "learning_rate": 6.802588996763755e-05,
1479
- "loss": 0.0381,
1480
- "step": 2090
1481
- },
1482
- {
1483
- "epoch": 6.688995215311005,
1484
- "grad_norm": 2.1549739837646484,
1485
- "learning_rate": 6.737864077669903e-05,
1486
- "loss": 0.0368,
1487
- "step": 2100
1488
- },
1489
- {
1490
- "epoch": 6.720893141945774,
1491
- "grad_norm": 1.404586672782898,
1492
- "learning_rate": 6.673139158576052e-05,
1493
- "loss": 0.0362,
1494
- "step": 2110
1495
- },
1496
- {
1497
- "epoch": 6.752791068580542,
1498
- "grad_norm": 2.1748902797698975,
1499
- "learning_rate": 6.608414239482201e-05,
1500
- "loss": 0.0358,
1501
- "step": 2120
1502
- },
1503
- {
1504
- "epoch": 6.784688995215311,
1505
- "grad_norm": 1.4457738399505615,
1506
- "learning_rate": 6.54368932038835e-05,
1507
- "loss": 0.0345,
1508
- "step": 2130
1509
- },
1510
- {
1511
- "epoch": 6.81658692185008,
1512
- "grad_norm": 1.1100033521652222,
1513
- "learning_rate": 6.478964401294499e-05,
1514
- "loss": 0.0329,
1515
- "step": 2140
1516
- },
1517
- {
1518
- "epoch": 6.848484848484849,
1519
- "grad_norm": 1.7228426933288574,
1520
- "learning_rate": 6.414239482200648e-05,
1521
- "loss": 0.0401,
1522
- "step": 2150
1523
- },
1524
- {
1525
- "epoch": 6.880382775119617,
1526
- "grad_norm": 1.428963541984558,
1527
- "learning_rate": 6.349514563106797e-05,
1528
- "loss": 0.0334,
1529
- "step": 2160
1530
- },
1531
- {
1532
- "epoch": 6.912280701754386,
1533
- "grad_norm": 0.96390700340271,
1534
- "learning_rate": 6.284789644012945e-05,
1535
- "loss": 0.0332,
1536
- "step": 2170
1537
- },
1538
- {
1539
- "epoch": 6.944178628389155,
1540
- "grad_norm": 0.9826375246047974,
1541
- "learning_rate": 6.220064724919095e-05,
1542
- "loss": 0.0348,
1543
- "step": 2180
1544
- },
1545
- {
1546
- "epoch": 6.976076555023924,
1547
- "grad_norm": 1.523210883140564,
1548
- "learning_rate": 6.155339805825243e-05,
1549
- "loss": 0.034,
1550
- "step": 2190
1551
- },
1552
- {
1553
- "epoch": 7.006379585326954,
1554
- "grad_norm": 0.9604802131652832,
1555
- "learning_rate": 6.090614886731392e-05,
1556
- "loss": 0.0324,
1557
- "step": 2200
1558
- },
1559
- {
1560
- "epoch": 7.038277511961723,
1561
- "grad_norm": 1.191280484199524,
1562
- "learning_rate": 6.02588996763754e-05,
1563
- "loss": 0.0296,
1564
- "step": 2210
1565
- },
1566
- {
1567
- "epoch": 7.0701754385964914,
1568
- "grad_norm": 0.9107991456985474,
1569
- "learning_rate": 5.96116504854369e-05,
1570
- "loss": 0.0295,
1571
- "step": 2220
1572
- },
1573
- {
1574
- "epoch": 7.10207336523126,
1575
- "grad_norm": 1.070676326751709,
1576
- "learning_rate": 5.8964401294498384e-05,
1577
- "loss": 0.0294,
1578
- "step": 2230
1579
- },
1580
- {
1581
- "epoch": 7.133971291866029,
1582
- "grad_norm": 1.4254604578018188,
1583
- "learning_rate": 5.831715210355987e-05,
1584
- "loss": 0.0333,
1585
- "step": 2240
1586
- },
1587
- {
1588
- "epoch": 7.165869218500798,
1589
- "grad_norm": 0.7672649025917053,
1590
- "learning_rate": 5.7669902912621365e-05,
1591
- "loss": 0.0283,
1592
- "step": 2250
1593
- },
1594
- {
1595
- "epoch": 7.1977671451355665,
1596
- "grad_norm": 1.1972182989120483,
1597
- "learning_rate": 5.702265372168285e-05,
1598
- "loss": 0.0316,
1599
- "step": 2260
1600
- },
1601
- {
1602
- "epoch": 7.229665071770335,
1603
- "grad_norm": 0.9965393543243408,
1604
- "learning_rate": 5.637540453074433e-05,
1605
- "loss": 0.0295,
1606
- "step": 2270
1607
- },
1608
- {
1609
- "epoch": 7.261562998405104,
1610
- "grad_norm": 0.7158511281013489,
1611
- "learning_rate": 5.572815533980583e-05,
1612
- "loss": 0.0285,
1613
- "step": 2280
1614
- },
1615
- {
1616
- "epoch": 7.293460925039873,
1617
- "grad_norm": 0.9511866569519043,
1618
- "learning_rate": 5.5080906148867314e-05,
1619
- "loss": 0.0313,
1620
- "step": 2290
1621
- },
1622
- {
1623
- "epoch": 7.3253588516746415,
1624
- "grad_norm": 1.0649924278259277,
1625
- "learning_rate": 5.4433656957928805e-05,
1626
- "loss": 0.0291,
1627
- "step": 2300
1628
- },
1629
- {
1630
- "epoch": 7.35725677830941,
1631
- "grad_norm": 0.8110047578811646,
1632
- "learning_rate": 5.37864077669903e-05,
1633
- "loss": 0.027,
1634
- "step": 2310
1635
- },
1636
- {
1637
- "epoch": 7.389154704944179,
1638
- "grad_norm": 1.2479182481765747,
1639
- "learning_rate": 5.3139158576051786e-05,
1640
- "loss": 0.0284,
1641
- "step": 2320
1642
- },
1643
- {
1644
- "epoch": 7.421052631578947,
1645
- "grad_norm": 1.4915539026260376,
1646
- "learning_rate": 5.249190938511327e-05,
1647
- "loss": 0.0283,
1648
- "step": 2330
1649
- },
1650
- {
1651
- "epoch": 7.4529505582137165,
1652
- "grad_norm": 1.0583655834197998,
1653
- "learning_rate": 5.1844660194174753e-05,
1654
- "loss": 0.0274,
1655
- "step": 2340
1656
- },
1657
- {
1658
- "epoch": 7.484848484848484,
1659
- "grad_norm": 0.9848374128341675,
1660
- "learning_rate": 5.119741100323625e-05,
1661
- "loss": 0.0287,
1662
- "step": 2350
1663
- },
1664
- {
1665
- "epoch": 7.516746411483254,
1666
- "grad_norm": 0.8352569341659546,
1667
- "learning_rate": 5.0550161812297735e-05,
1668
- "loss": 0.0275,
1669
- "step": 2360
1670
- },
1671
- {
1672
- "epoch": 7.548644338118022,
1673
- "grad_norm": 0.756790816783905,
1674
- "learning_rate": 4.9902912621359225e-05,
1675
- "loss": 0.026,
1676
- "step": 2370
1677
- },
1678
- {
1679
- "epoch": 7.580542264752791,
1680
- "grad_norm": 0.7591552138328552,
1681
- "learning_rate": 4.925566343042071e-05,
1682
- "loss": 0.0271,
1683
- "step": 2380
1684
- },
1685
- {
1686
- "epoch": 7.6124401913875595,
1687
- "grad_norm": 0.6143433451652527,
1688
- "learning_rate": 4.86084142394822e-05,
1689
- "loss": 0.026,
1690
- "step": 2390
1691
- },
1692
- {
1693
- "epoch": 7.644338118022328,
1694
- "grad_norm": 1.1552224159240723,
1695
- "learning_rate": 4.796116504854369e-05,
1696
- "loss": 0.0258,
1697
- "step": 2400
1698
- },
1699
- {
1700
- "epoch": 7.676236044657097,
1701
- "grad_norm": 0.9515504837036133,
1702
- "learning_rate": 4.731391585760518e-05,
1703
- "loss": 0.0248,
1704
- "step": 2410
1705
- },
1706
- {
1707
- "epoch": 7.708133971291866,
1708
- "grad_norm": 0.4848612844944,
1709
- "learning_rate": 4.666666666666667e-05,
1710
- "loss": 0.0257,
1711
- "step": 2420
1712
- },
1713
- {
1714
- "epoch": 7.7400318979266345,
1715
- "grad_norm": 0.7792959809303284,
1716
- "learning_rate": 4.6019417475728155e-05,
1717
- "loss": 0.0247,
1718
- "step": 2430
1719
- },
1720
- {
1721
- "epoch": 7.771929824561403,
1722
- "grad_norm": 0.798723042011261,
1723
- "learning_rate": 4.5372168284789646e-05,
1724
- "loss": 0.0243,
1725
- "step": 2440
1726
- },
1727
- {
1728
- "epoch": 7.803827751196172,
1729
- "grad_norm": 0.38535213470458984,
1730
- "learning_rate": 4.472491909385114e-05,
1731
- "loss": 0.024,
1732
- "step": 2450
1733
- },
1734
- {
1735
- "epoch": 7.835725677830941,
1736
- "grad_norm": 0.7133737802505493,
1737
- "learning_rate": 4.407766990291262e-05,
1738
- "loss": 0.0239,
1739
- "step": 2460
1740
- },
1741
- {
1742
- "epoch": 7.8676236044657095,
1743
- "grad_norm": 0.6840415596961975,
1744
- "learning_rate": 4.343042071197411e-05,
1745
- "loss": 0.0239,
1746
- "step": 2470
1747
- },
1748
- {
1749
- "epoch": 7.899521531100478,
1750
- "grad_norm": 0.5434433817863464,
1751
- "learning_rate": 4.27831715210356e-05,
1752
- "loss": 0.0232,
1753
- "step": 2480
1754
- },
1755
- {
1756
- "epoch": 7.931419457735247,
1757
- "grad_norm": 0.42021751403808594,
1758
- "learning_rate": 4.2135922330097086e-05,
1759
- "loss": 0.0231,
1760
- "step": 2490
1761
- },
1762
- {
1763
- "epoch": 7.963317384370016,
1764
- "grad_norm": 0.24440300464630127,
1765
- "learning_rate": 4.1488673139158576e-05,
1766
- "loss": 0.0221,
1767
- "step": 2500
1768
- },
1769
- {
1770
- "epoch": 7.9952153110047846,
1771
- "grad_norm": 0.32175081968307495,
1772
- "learning_rate": 4.084142394822007e-05,
1773
- "loss": 0.0215,
1774
- "step": 2510
1775
- },
1776
- {
1777
- "epoch": 8.025518341307816,
1778
- "grad_norm": 0.26291289925575256,
1779
- "learning_rate": 4.019417475728156e-05,
1780
- "loss": 0.0204,
1781
- "step": 2520
1782
- },
1783
- {
1784
- "epoch": 8.057416267942584,
1785
- "grad_norm": 0.27921006083488464,
1786
- "learning_rate": 3.954692556634305e-05,
1787
- "loss": 0.0202,
1788
- "step": 2530
1789
- },
1790
- {
1791
- "epoch": 8.089314194577353,
1792
- "grad_norm": 0.4434060752391815,
1793
- "learning_rate": 3.889967637540453e-05,
1794
- "loss": 0.0204,
1795
- "step": 2540
1796
- },
1797
- {
1798
- "epoch": 8.121212121212121,
1799
- "grad_norm": 0.3078613877296448,
1800
- "learning_rate": 3.825242718446602e-05,
1801
- "loss": 0.0208,
1802
- "step": 2550
1803
- },
1804
- {
1805
- "epoch": 8.15311004784689,
1806
- "grad_norm": 0.23100249469280243,
1807
- "learning_rate": 3.7605177993527506e-05,
1808
- "loss": 0.0207,
1809
- "step": 2560
1810
- },
1811
- {
1812
- "epoch": 8.185007974481659,
1813
- "grad_norm": 0.34067490696907043,
1814
- "learning_rate": 3.6957928802589e-05,
1815
- "loss": 0.0207,
1816
- "step": 2570
1817
- },
1818
- {
1819
- "epoch": 8.216905901116428,
1820
- "grad_norm": 0.30456990003585815,
1821
- "learning_rate": 3.631067961165049e-05,
1822
- "loss": 0.0215,
1823
- "step": 2580
1824
- },
1825
- {
1826
- "epoch": 8.248803827751196,
1827
- "grad_norm": 0.27407306432724,
1828
- "learning_rate": 3.566343042071197e-05,
1829
- "loss": 0.02,
1830
- "step": 2590
1831
- },
1832
- {
1833
- "epoch": 8.280701754385966,
1834
- "grad_norm": 0.3087022304534912,
1835
- "learning_rate": 3.501618122977346e-05,
1836
- "loss": 0.0206,
1837
- "step": 2600
1838
- },
1839
- {
1840
- "epoch": 8.312599681020734,
1841
- "grad_norm": 0.5118716359138489,
1842
- "learning_rate": 3.436893203883495e-05,
1843
- "loss": 0.0207,
1844
- "step": 2610
1845
- },
1846
- {
1847
- "epoch": 8.344497607655502,
1848
- "grad_norm": 0.2015773355960846,
1849
- "learning_rate": 3.372168284789644e-05,
1850
- "loss": 0.02,
1851
- "step": 2620
1852
- },
1853
- {
1854
- "epoch": 8.376395534290271,
1855
- "grad_norm": 0.2432379424571991,
1856
- "learning_rate": 3.3074433656957934e-05,
1857
- "loss": 0.0201,
1858
- "step": 2630
1859
- },
1860
- {
1861
- "epoch": 8.408293460925039,
1862
- "grad_norm": 0.2935405373573303,
1863
- "learning_rate": 3.242718446601942e-05,
1864
- "loss": 0.0209,
1865
- "step": 2640
1866
- },
1867
- {
1868
- "epoch": 8.440191387559809,
1869
- "grad_norm": 0.2740592658519745,
1870
- "learning_rate": 3.177993527508091e-05,
1871
- "loss": 0.0204,
1872
- "step": 2650
1873
- },
1874
- {
1875
- "epoch": 8.472089314194577,
1876
- "grad_norm": 0.19506220519542694,
1877
- "learning_rate": 3.11326860841424e-05,
1878
- "loss": 0.0204,
1879
- "step": 2660
1880
- },
1881
- {
1882
- "epoch": 8.503987240829346,
1883
- "grad_norm": 0.2191346436738968,
1884
- "learning_rate": 3.0485436893203883e-05,
1885
- "loss": 0.02,
1886
- "step": 2670
1887
- },
1888
- {
1889
- "epoch": 8.535885167464114,
1890
- "grad_norm": 0.17195333540439606,
1891
- "learning_rate": 2.9838187702265373e-05,
1892
- "loss": 0.0203,
1893
- "step": 2680
1894
- },
1895
- {
1896
- "epoch": 8.567783094098884,
1897
- "grad_norm": 0.1918991059064865,
1898
- "learning_rate": 2.919093851132686e-05,
1899
- "loss": 0.0199,
1900
- "step": 2690
1901
- },
1902
- {
1903
- "epoch": 8.599681020733652,
1904
- "grad_norm": 0.32339102029800415,
1905
- "learning_rate": 2.854368932038835e-05,
1906
- "loss": 0.0207,
1907
- "step": 2700
1908
- },
1909
- {
1910
- "epoch": 8.631578947368421,
1911
- "grad_norm": 0.21508780121803284,
1912
- "learning_rate": 2.7896440129449842e-05,
1913
- "loss": 0.0206,
1914
- "step": 2710
1915
- },
1916
- {
1917
- "epoch": 8.663476874003189,
1918
- "grad_norm": 0.186102032661438,
1919
- "learning_rate": 2.7249190938511326e-05,
1920
- "loss": 0.0204,
1921
- "step": 2720
1922
- },
1923
- {
1924
- "epoch": 8.695374800637959,
1925
- "grad_norm": 0.44521018862724304,
1926
- "learning_rate": 2.6601941747572816e-05,
1927
- "loss": 0.0195,
1928
- "step": 2730
1929
- },
1930
- {
1931
- "epoch": 8.727272727272727,
1932
- "grad_norm": 0.2109213024377823,
1933
- "learning_rate": 2.5954692556634307e-05,
1934
- "loss": 0.0198,
1935
- "step": 2740
1936
- },
1937
- {
1938
- "epoch": 8.759170653907496,
1939
- "grad_norm": 0.19141490757465363,
1940
- "learning_rate": 2.5307443365695794e-05,
1941
- "loss": 0.02,
1942
- "step": 2750
1943
- },
1944
- {
1945
- "epoch": 8.791068580542264,
1946
- "grad_norm": 0.19844338297843933,
1947
- "learning_rate": 2.4660194174757285e-05,
1948
- "loss": 0.0203,
1949
- "step": 2760
1950
- },
1951
- {
1952
- "epoch": 8.822966507177034,
1953
- "grad_norm": 0.22535575926303864,
1954
- "learning_rate": 2.4012944983818772e-05,
1955
- "loss": 0.0202,
1956
- "step": 2770
1957
- },
1958
- {
1959
- "epoch": 8.854864433811802,
1960
- "grad_norm": 0.15982511639595032,
1961
- "learning_rate": 2.336569579288026e-05,
1962
- "loss": 0.0195,
1963
- "step": 2780
1964
- },
1965
- {
1966
- "epoch": 8.886762360446571,
1967
- "grad_norm": 0.34473344683647156,
1968
- "learning_rate": 2.2718446601941746e-05,
1969
- "loss": 0.02,
1970
- "step": 2790
1971
- },
1972
- {
1973
- "epoch": 8.91866028708134,
1974
- "grad_norm": 0.2416442632675171,
1975
- "learning_rate": 2.2071197411003237e-05,
1976
- "loss": 0.0199,
1977
- "step": 2800
1978
- },
1979
- {
1980
- "epoch": 8.950558213716109,
1981
- "grad_norm": 0.32778188586235046,
1982
- "learning_rate": 2.1423948220064728e-05,
1983
- "loss": 0.0201,
1984
- "step": 2810
1985
- },
1986
- {
1987
- "epoch": 8.982456140350877,
1988
- "grad_norm": 0.18563535809516907,
1989
- "learning_rate": 2.0776699029126215e-05,
1990
- "loss": 0.0201,
1991
- "step": 2820
1992
- },
1993
- {
1994
- "epoch": 9.012759170653908,
1995
- "grad_norm": 0.2312159687280655,
1996
- "learning_rate": 2.0129449838187702e-05,
1997
- "loss": 0.019,
1998
- "step": 2830
1999
- },
2000
- {
2001
- "epoch": 9.044657097288676,
2002
- "grad_norm": 0.22602292895317078,
2003
- "learning_rate": 1.948220064724919e-05,
2004
- "loss": 0.0197,
2005
- "step": 2840
2006
- },
2007
- {
2008
- "epoch": 9.076555023923445,
2009
- "grad_norm": 0.2483113408088684,
2010
- "learning_rate": 1.883495145631068e-05,
2011
- "loss": 0.0191,
2012
- "step": 2850
2013
- },
2014
- {
2015
- "epoch": 9.108452950558213,
2016
- "grad_norm": 0.19213229417800903,
2017
- "learning_rate": 1.818770226537217e-05,
2018
- "loss": 0.019,
2019
- "step": 2860
2020
- },
2021
- {
2022
- "epoch": 9.140350877192983,
2023
- "grad_norm": 0.20552313327789307,
2024
- "learning_rate": 1.7540453074433658e-05,
2025
- "loss": 0.0196,
2026
- "step": 2870
2027
- },
2028
- {
2029
- "epoch": 9.17224880382775,
2030
- "grad_norm": 0.1715112179517746,
2031
- "learning_rate": 1.6893203883495145e-05,
2032
- "loss": 0.0199,
2033
- "step": 2880
2034
- },
2035
- {
2036
- "epoch": 9.20414673046252,
2037
- "grad_norm": 0.1956663876771927,
2038
- "learning_rate": 1.6245954692556636e-05,
2039
- "loss": 0.0192,
2040
- "step": 2890
2041
- },
2042
- {
2043
- "epoch": 9.236044657097288,
2044
- "grad_norm": 0.20990578830242157,
2045
- "learning_rate": 1.5598705501618123e-05,
2046
- "loss": 0.0192,
2047
- "step": 2900
2048
- },
2049
- {
2050
- "epoch": 9.267942583732058,
2051
- "grad_norm": 0.21641728281974792,
2052
- "learning_rate": 1.4951456310679612e-05,
2053
- "loss": 0.0194,
2054
- "step": 2910
2055
- },
2056
- {
2057
- "epoch": 9.299840510366826,
2058
- "grad_norm": 0.148260235786438,
2059
- "learning_rate": 1.43042071197411e-05,
2060
- "loss": 0.0197,
2061
- "step": 2920
2062
- },
2063
- {
2064
- "epoch": 9.331738437001595,
2065
- "grad_norm": 0.24010303616523743,
2066
- "learning_rate": 1.3656957928802588e-05,
2067
- "loss": 0.0194,
2068
- "step": 2930
2069
- },
2070
- {
2071
- "epoch": 9.363636363636363,
2072
- "grad_norm": 0.18212716281414032,
2073
- "learning_rate": 1.3009708737864079e-05,
2074
- "loss": 0.0192,
2075
- "step": 2940
2076
- },
2077
- {
2078
- "epoch": 9.395534290271133,
2079
- "grad_norm": 0.18527287244796753,
2080
- "learning_rate": 1.2362459546925568e-05,
2081
- "loss": 0.02,
2082
- "step": 2950
2083
- },
2084
- {
2085
- "epoch": 9.4274322169059,
2086
- "grad_norm": 0.16474080085754395,
2087
- "learning_rate": 1.1715210355987055e-05,
2088
- "loss": 0.0193,
2089
- "step": 2960
2090
- },
2091
- {
2092
- "epoch": 9.45933014354067,
2093
- "grad_norm": 0.16805745661258698,
2094
- "learning_rate": 1.1067961165048545e-05,
2095
- "loss": 0.0189,
2096
- "step": 2970
2097
- },
2098
- {
2099
- "epoch": 9.491228070175438,
2100
- "grad_norm": 0.20816102623939514,
2101
- "learning_rate": 1.0420711974110033e-05,
2102
- "loss": 0.0193,
2103
- "step": 2980
2104
- },
2105
- {
2106
- "epoch": 9.523125996810208,
2107
- "grad_norm": 0.1942005753517151,
2108
- "learning_rate": 9.773462783171522e-06,
2109
- "loss": 0.0191,
2110
- "step": 2990
2111
- },
2112
- {
2113
- "epoch": 9.555023923444976,
2114
- "grad_norm": 0.17980408668518066,
2115
- "learning_rate": 9.12621359223301e-06,
2116
- "loss": 0.0187,
2117
- "step": 3000
2118
- },
2119
- {
2120
- "epoch": 9.586921850079746,
2121
- "grad_norm": 0.2269669771194458,
2122
- "learning_rate": 8.4789644012945e-06,
2123
- "loss": 0.0193,
2124
- "step": 3010
2125
- },
2126
- {
2127
- "epoch": 9.618819776714513,
2128
- "grad_norm": 0.2470860481262207,
2129
- "learning_rate": 7.831715210355987e-06,
2130
- "loss": 0.0197,
2131
- "step": 3020
2132
- },
2133
- {
2134
- "epoch": 9.650717703349283,
2135
- "grad_norm": 0.18939535319805145,
2136
- "learning_rate": 7.1844660194174755e-06,
2137
- "loss": 0.0195,
2138
- "step": 3030
2139
- },
2140
- {
2141
- "epoch": 9.682615629984051,
2142
- "grad_norm": 0.2230576127767563,
2143
- "learning_rate": 6.5372168284789644e-06,
2144
- "loss": 0.0196,
2145
- "step": 3040
2146
- },
2147
- {
2148
- "epoch": 9.71451355661882,
2149
- "grad_norm": 0.22115571796894073,
2150
- "learning_rate": 5.889967637540453e-06,
2151
- "loss": 0.0193,
2152
- "step": 3050
2153
- },
2154
- {
2155
- "epoch": 9.746411483253588,
2156
- "grad_norm": 0.22138074040412903,
2157
- "learning_rate": 5.242718446601942e-06,
2158
- "loss": 0.0188,
2159
- "step": 3060
2160
- },
2161
- {
2162
- "epoch": 9.778309409888358,
2163
- "grad_norm": 0.17522794008255005,
2164
- "learning_rate": 4.595469255663431e-06,
2165
- "loss": 0.0192,
2166
- "step": 3070
2167
- },
2168
- {
2169
- "epoch": 9.810207336523126,
2170
- "grad_norm": 0.22353076934814453,
2171
- "learning_rate": 3.948220064724919e-06,
2172
- "loss": 0.019,
2173
- "step": 3080
2174
- },
2175
- {
2176
- "epoch": 9.842105263157894,
2177
- "grad_norm": 0.17822624742984772,
2178
- "learning_rate": 3.300970873786408e-06,
2179
- "loss": 0.0191,
2180
- "step": 3090
2181
- },
2182
- {
2183
- "epoch": 9.874003189792663,
2184
- "grad_norm": 0.18609146773815155,
2185
- "learning_rate": 2.6537216828478967e-06,
2186
- "loss": 0.0194,
2187
- "step": 3100
2188
- },
2189
- {
2190
- "epoch": 9.905901116427433,
2191
- "grad_norm": 0.16276702284812927,
2192
- "learning_rate": 2.006472491909385e-06,
2193
- "loss": 0.0193,
2194
- "step": 3110
2195
- },
2196
- {
2197
- "epoch": 9.937799043062201,
2198
- "grad_norm": 0.2207743376493454,
2199
- "learning_rate": 1.359223300970874e-06,
2200
- "loss": 0.0192,
2201
- "step": 3120
2202
- },
2203
- {
2204
- "epoch": 9.969696969696969,
2205
- "grad_norm": 0.1925983428955078,
2206
- "learning_rate": 7.119741100323625e-07,
2207
- "loss": 0.0193,
2208
- "step": 3130
2209
- },
2210
- {
2211
- "epoch": 10.0,
2212
- "grad_norm": 0.26687198877334595,
2213
- "learning_rate": 6.472491909385113e-08,
2214
- "loss": 0.0182,
2215
- "step": 3140
2216
- }
2217
- ],
2218
- "logging_steps": 10,
2219
- "max_steps": 3140,
2220
- "num_input_tokens_seen": 0,
2221
- "num_train_epochs": 10,
2222
- "save_steps": 500,
2223
- "stateful_callbacks": {
2224
- "TrainerControl": {
2225
- "args": {
2226
- "should_epoch_stop": false,
2227
- "should_evaluate": false,
2228
- "should_log": false,
2229
- "should_save": true,
2230
- "should_training_stop": true
2231
- },
2232
- "attributes": {}
2233
- }
2234
- },
2235
- "total_flos": 2.344970104917197e+18,
2236
- "train_batch_size": 2,
2237
- "trial_name": null,
2238
- "trial_params": null
2239
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
checkpoints/training_args.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:447402a79d5f5069eae8b80718807636c92dad555b106350f5427516264d4b1b
3
- size 5841