Dorjzodovsuren commited on
Commit
3f23b04
1 Parent(s): cc5ea79

Upload 12 files

Browse files

the last checkpoint

generation_config.json ADDED
@@ -0,0 +1,263 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 5,
5
+ 3
6
+ ],
7
+ [
8
+ 5,
9
+ 9
10
+ ],
11
+ [
12
+ 8,
13
+ 0
14
+ ],
15
+ [
16
+ 8,
17
+ 4
18
+ ],
19
+ [
20
+ 8,
21
+ 7
22
+ ],
23
+ [
24
+ 8,
25
+ 8
26
+ ],
27
+ [
28
+ 9,
29
+ 0
30
+ ],
31
+ [
32
+ 9,
33
+ 7
34
+ ],
35
+ [
36
+ 9,
37
+ 9
38
+ ],
39
+ [
40
+ 10,
41
+ 5
42
+ ]
43
+ ],
44
+ "begin_suppress_tokens": [
45
+ 220,
46
+ 50257
47
+ ],
48
+ "bos_token_id": 50257,
49
+ "decoder_start_token_id": 50258,
50
+ "eos_token_id": 50257,
51
+ "forced_decoder_ids": [
52
+ [
53
+ 1,
54
+ null
55
+ ],
56
+ [
57
+ 2,
58
+ 50359
59
+ ]
60
+ ],
61
+ "is_multilingual": true,
62
+ "lang_to_id": {
63
+ "<|af|>": 50327,
64
+ "<|am|>": 50334,
65
+ "<|ar|>": 50272,
66
+ "<|as|>": 50350,
67
+ "<|az|>": 50304,
68
+ "<|ba|>": 50355,
69
+ "<|be|>": 50330,
70
+ "<|bg|>": 50292,
71
+ "<|bn|>": 50302,
72
+ "<|bo|>": 50347,
73
+ "<|br|>": 50309,
74
+ "<|bs|>": 50315,
75
+ "<|ca|>": 50270,
76
+ "<|cs|>": 50283,
77
+ "<|cy|>": 50297,
78
+ "<|da|>": 50285,
79
+ "<|de|>": 50261,
80
+ "<|el|>": 50281,
81
+ "<|en|>": 50259,
82
+ "<|es|>": 50262,
83
+ "<|et|>": 50307,
84
+ "<|eu|>": 50310,
85
+ "<|fa|>": 50300,
86
+ "<|fi|>": 50277,
87
+ "<|fo|>": 50338,
88
+ "<|fr|>": 50265,
89
+ "<|gl|>": 50319,
90
+ "<|gu|>": 50333,
91
+ "<|haw|>": 50352,
92
+ "<|ha|>": 50354,
93
+ "<|he|>": 50279,
94
+ "<|hi|>": 50276,
95
+ "<|hr|>": 50291,
96
+ "<|ht|>": 50339,
97
+ "<|hu|>": 50286,
98
+ "<|hy|>": 50312,
99
+ "<|id|>": 50275,
100
+ "<|is|>": 50311,
101
+ "<|it|>": 50274,
102
+ "<|ja|>": 50266,
103
+ "<|jw|>": 50356,
104
+ "<|ka|>": 50329,
105
+ "<|kk|>": 50316,
106
+ "<|km|>": 50323,
107
+ "<|kn|>": 50306,
108
+ "<|ko|>": 50264,
109
+ "<|la|>": 50294,
110
+ "<|lb|>": 50345,
111
+ "<|ln|>": 50353,
112
+ "<|lo|>": 50336,
113
+ "<|lt|>": 50293,
114
+ "<|lv|>": 50301,
115
+ "<|mg|>": 50349,
116
+ "<|mi|>": 50295,
117
+ "<|mk|>": 50308,
118
+ "<|ml|>": 50296,
119
+ "<|mn|>": 50314,
120
+ "<|mr|>": 50320,
121
+ "<|ms|>": 50282,
122
+ "<|mt|>": 50343,
123
+ "<|my|>": 50346,
124
+ "<|ne|>": 50313,
125
+ "<|nl|>": 50271,
126
+ "<|nn|>": 50342,
127
+ "<|no|>": 50288,
128
+ "<|oc|>": 50328,
129
+ "<|pa|>": 50321,
130
+ "<|pl|>": 50269,
131
+ "<|ps|>": 50340,
132
+ "<|pt|>": 50267,
133
+ "<|ro|>": 50284,
134
+ "<|ru|>": 50263,
135
+ "<|sa|>": 50344,
136
+ "<|sd|>": 50332,
137
+ "<|si|>": 50322,
138
+ "<|sk|>": 50298,
139
+ "<|sl|>": 50305,
140
+ "<|sn|>": 50324,
141
+ "<|so|>": 50326,
142
+ "<|sq|>": 50317,
143
+ "<|sr|>": 50303,
144
+ "<|su|>": 50357,
145
+ "<|sv|>": 50273,
146
+ "<|sw|>": 50318,
147
+ "<|ta|>": 50287,
148
+ "<|te|>": 50299,
149
+ "<|tg|>": 50331,
150
+ "<|th|>": 50289,
151
+ "<|tk|>": 50341,
152
+ "<|tl|>": 50348,
153
+ "<|tr|>": 50268,
154
+ "<|tt|>": 50351,
155
+ "<|uk|>": 50280,
156
+ "<|ur|>": 50290,
157
+ "<|uz|>": 50337,
158
+ "<|vi|>": 50278,
159
+ "<|yi|>": 50335,
160
+ "<|yo|>": 50325,
161
+ "<|zh|>": 50260
162
+ },
163
+ "max_initial_timestamp_index": 1,
164
+ "max_length": 448,
165
+ "no_timestamps_token_id": 50363,
166
+ "pad_token_id": 50257,
167
+ "return_timestamps": false,
168
+ "suppress_tokens": [
169
+ 1,
170
+ 2,
171
+ 7,
172
+ 8,
173
+ 9,
174
+ 10,
175
+ 14,
176
+ 25,
177
+ 26,
178
+ 27,
179
+ 28,
180
+ 29,
181
+ 31,
182
+ 58,
183
+ 59,
184
+ 60,
185
+ 61,
186
+ 62,
187
+ 63,
188
+ 90,
189
+ 91,
190
+ 92,
191
+ 93,
192
+ 359,
193
+ 503,
194
+ 522,
195
+ 542,
196
+ 873,
197
+ 893,
198
+ 902,
199
+ 918,
200
+ 922,
201
+ 931,
202
+ 1350,
203
+ 1853,
204
+ 1982,
205
+ 2460,
206
+ 2627,
207
+ 3246,
208
+ 3253,
209
+ 3268,
210
+ 3536,
211
+ 3846,
212
+ 3961,
213
+ 4183,
214
+ 4667,
215
+ 6585,
216
+ 6647,
217
+ 7273,
218
+ 9061,
219
+ 9383,
220
+ 10428,
221
+ 10929,
222
+ 11938,
223
+ 12033,
224
+ 12331,
225
+ 12562,
226
+ 13793,
227
+ 14157,
228
+ 14635,
229
+ 15265,
230
+ 15618,
231
+ 16553,
232
+ 16604,
233
+ 18362,
234
+ 18956,
235
+ 20075,
236
+ 21675,
237
+ 22520,
238
+ 26130,
239
+ 26161,
240
+ 26435,
241
+ 28279,
242
+ 29464,
243
+ 31650,
244
+ 32302,
245
+ 32470,
246
+ 36865,
247
+ 42863,
248
+ 47425,
249
+ 49870,
250
+ 50254,
251
+ 50258,
252
+ 50358,
253
+ 50359,
254
+ 50360,
255
+ 50361,
256
+ 50362
257
+ ],
258
+ "task_to_id": {
259
+ "transcribe": 50359,
260
+ "translate": 50358
261
+ },
262
+ "transformers_version": "4.36.2"
263
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:387fec55a8b05d1d68bef3700e646bd520f5df859de683d732831050714b6e7e
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ead66f31896f988c1667a5eb83027ee2d52c2c170f11d39de53a7d2aeeb97c9
3
  size 966995080
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e521a355887780be4566d823bff9b4878173fbe276d85729d6ccb366f26ac03d
3
+ size 1925064044
rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fd0829446aed2a22a6b73293fd9b28397dd5cc9178ad560f905345507e0b14c
3
+ size 15024
rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d33d234f3340e2e02e73034c9d410e94662bb79c8463af61eb95952816955065
3
+ size 15024
rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1722e237813140b13d22493a3a95b8308c81b7e0c12fbf0642ddb52ae3428c72
3
+ size 15024
rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b178bd00126cf6069df039a0145c3e0edff8d185ead2392f27b1ceffa2d73c1b
3
+ size 15024
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7efbf80617c96c78286826ce59d9a12c86da62d7631874b3d6364a8e993ada60
3
+ size 1064
trainer_state.json ADDED
@@ -0,0 +1,1017 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 52.89689034369886,
3
+ "best_model_checkpoint": "./whisper-small_v2_mn/checkpoint-2000",
4
+ "epoch": 31.746031746031747,
5
+ "eval_steps": 1000,
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": 0.2,
13
+ "learning_rate": 4.2000000000000006e-07,
14
+ "loss": 5.8081,
15
+ "step": 25
16
+ },
17
+ {
18
+ "epoch": 0.4,
19
+ "learning_rate": 9.200000000000001e-07,
20
+ "loss": 4.3345,
21
+ "step": 50
22
+ },
23
+ {
24
+ "epoch": 0.6,
25
+ "learning_rate": 1.42e-06,
26
+ "loss": 3.1424,
27
+ "step": 75
28
+ },
29
+ {
30
+ "epoch": 0.79,
31
+ "learning_rate": 1.9200000000000003e-06,
32
+ "loss": 2.3848,
33
+ "step": 100
34
+ },
35
+ {
36
+ "epoch": 0.99,
37
+ "learning_rate": 2.42e-06,
38
+ "loss": 1.8362,
39
+ "step": 125
40
+ },
41
+ {
42
+ "epoch": 1.19,
43
+ "learning_rate": 2.92e-06,
44
+ "loss": 1.4852,
45
+ "step": 150
46
+ },
47
+ {
48
+ "epoch": 1.39,
49
+ "learning_rate": 3.4200000000000007e-06,
50
+ "loss": 1.2819,
51
+ "step": 175
52
+ },
53
+ {
54
+ "epoch": 1.59,
55
+ "learning_rate": 3.920000000000001e-06,
56
+ "loss": 1.1295,
57
+ "step": 200
58
+ },
59
+ {
60
+ "epoch": 1.79,
61
+ "learning_rate": 4.42e-06,
62
+ "loss": 0.9959,
63
+ "step": 225
64
+ },
65
+ {
66
+ "epoch": 1.98,
67
+ "learning_rate": 4.92e-06,
68
+ "loss": 0.9184,
69
+ "step": 250
70
+ },
71
+ {
72
+ "epoch": 2.18,
73
+ "learning_rate": 5.420000000000001e-06,
74
+ "loss": 0.7924,
75
+ "step": 275
76
+ },
77
+ {
78
+ "epoch": 2.38,
79
+ "learning_rate": 5.92e-06,
80
+ "loss": 0.6941,
81
+ "step": 300
82
+ },
83
+ {
84
+ "epoch": 2.58,
85
+ "learning_rate": 6.42e-06,
86
+ "loss": 0.6417,
87
+ "step": 325
88
+ },
89
+ {
90
+ "epoch": 2.78,
91
+ "learning_rate": 6.92e-06,
92
+ "loss": 0.5299,
93
+ "step": 350
94
+ },
95
+ {
96
+ "epoch": 2.98,
97
+ "learning_rate": 7.420000000000001e-06,
98
+ "loss": 0.4726,
99
+ "step": 375
100
+ },
101
+ {
102
+ "epoch": 3.17,
103
+ "learning_rate": 7.92e-06,
104
+ "loss": 0.3759,
105
+ "step": 400
106
+ },
107
+ {
108
+ "epoch": 3.37,
109
+ "learning_rate": 8.42e-06,
110
+ "loss": 0.3508,
111
+ "step": 425
112
+ },
113
+ {
114
+ "epoch": 3.57,
115
+ "learning_rate": 8.920000000000001e-06,
116
+ "loss": 0.3384,
117
+ "step": 450
118
+ },
119
+ {
120
+ "epoch": 3.77,
121
+ "learning_rate": 9.42e-06,
122
+ "loss": 0.3153,
123
+ "step": 475
124
+ },
125
+ {
126
+ "epoch": 3.97,
127
+ "learning_rate": 9.920000000000002e-06,
128
+ "loss": 0.3337,
129
+ "step": 500
130
+ },
131
+ {
132
+ "epoch": 4.17,
133
+ "learning_rate": 9.940000000000001e-06,
134
+ "loss": 0.2261,
135
+ "step": 525
136
+ },
137
+ {
138
+ "epoch": 4.37,
139
+ "learning_rate": 9.86857142857143e-06,
140
+ "loss": 0.1987,
141
+ "step": 550
142
+ },
143
+ {
144
+ "epoch": 4.56,
145
+ "learning_rate": 9.797142857142858e-06,
146
+ "loss": 0.1959,
147
+ "step": 575
148
+ },
149
+ {
150
+ "epoch": 4.76,
151
+ "learning_rate": 9.725714285714287e-06,
152
+ "loss": 0.1938,
153
+ "step": 600
154
+ },
155
+ {
156
+ "epoch": 4.96,
157
+ "learning_rate": 9.654285714285716e-06,
158
+ "loss": 0.2021,
159
+ "step": 625
160
+ },
161
+ {
162
+ "epoch": 5.16,
163
+ "learning_rate": 9.582857142857143e-06,
164
+ "loss": 0.1245,
165
+ "step": 650
166
+ },
167
+ {
168
+ "epoch": 5.36,
169
+ "learning_rate": 9.511428571428572e-06,
170
+ "loss": 0.1071,
171
+ "step": 675
172
+ },
173
+ {
174
+ "epoch": 5.56,
175
+ "learning_rate": 9.440000000000001e-06,
176
+ "loss": 0.1053,
177
+ "step": 700
178
+ },
179
+ {
180
+ "epoch": 5.75,
181
+ "learning_rate": 9.368571428571428e-06,
182
+ "loss": 0.1179,
183
+ "step": 725
184
+ },
185
+ {
186
+ "epoch": 5.95,
187
+ "learning_rate": 9.297142857142857e-06,
188
+ "loss": 0.1051,
189
+ "step": 750
190
+ },
191
+ {
192
+ "epoch": 6.15,
193
+ "learning_rate": 9.225714285714286e-06,
194
+ "loss": 0.0661,
195
+ "step": 775
196
+ },
197
+ {
198
+ "epoch": 6.35,
199
+ "learning_rate": 9.154285714285715e-06,
200
+ "loss": 0.0528,
201
+ "step": 800
202
+ },
203
+ {
204
+ "epoch": 6.55,
205
+ "learning_rate": 9.082857142857143e-06,
206
+ "loss": 0.0578,
207
+ "step": 825
208
+ },
209
+ {
210
+ "epoch": 6.75,
211
+ "learning_rate": 9.011428571428572e-06,
212
+ "loss": 0.0566,
213
+ "step": 850
214
+ },
215
+ {
216
+ "epoch": 6.94,
217
+ "learning_rate": 8.94e-06,
218
+ "loss": 0.0568,
219
+ "step": 875
220
+ },
221
+ {
222
+ "epoch": 7.14,
223
+ "learning_rate": 8.86857142857143e-06,
224
+ "loss": 0.0395,
225
+ "step": 900
226
+ },
227
+ {
228
+ "epoch": 7.34,
229
+ "learning_rate": 8.797142857142857e-06,
230
+ "loss": 0.0313,
231
+ "step": 925
232
+ },
233
+ {
234
+ "epoch": 7.54,
235
+ "learning_rate": 8.725714285714286e-06,
236
+ "loss": 0.0302,
237
+ "step": 950
238
+ },
239
+ {
240
+ "epoch": 7.74,
241
+ "learning_rate": 8.654285714285715e-06,
242
+ "loss": 0.0309,
243
+ "step": 975
244
+ },
245
+ {
246
+ "epoch": 7.94,
247
+ "learning_rate": 8.582857142857144e-06,
248
+ "loss": 0.0312,
249
+ "step": 1000
250
+ },
251
+ {
252
+ "epoch": 7.94,
253
+ "eval_loss": 0.640009880065918,
254
+ "eval_runtime": 477.2063,
255
+ "eval_samples_per_second": 3.952,
256
+ "eval_steps_per_second": 0.124,
257
+ "eval_wer": 54.06983087834152,
258
+ "step": 1000
259
+ },
260
+ {
261
+ "epoch": 8.13,
262
+ "learning_rate": 8.511428571428571e-06,
263
+ "loss": 0.0234,
264
+ "step": 1025
265
+ },
266
+ {
267
+ "epoch": 8.33,
268
+ "learning_rate": 8.44e-06,
269
+ "loss": 0.0184,
270
+ "step": 1050
271
+ },
272
+ {
273
+ "epoch": 8.53,
274
+ "learning_rate": 8.36857142857143e-06,
275
+ "loss": 0.0178,
276
+ "step": 1075
277
+ },
278
+ {
279
+ "epoch": 8.73,
280
+ "learning_rate": 8.297142857142859e-06,
281
+ "loss": 0.0199,
282
+ "step": 1100
283
+ },
284
+ {
285
+ "epoch": 8.93,
286
+ "learning_rate": 8.225714285714288e-06,
287
+ "loss": 0.0173,
288
+ "step": 1125
289
+ },
290
+ {
291
+ "epoch": 9.13,
292
+ "learning_rate": 8.154285714285715e-06,
293
+ "loss": 0.0131,
294
+ "step": 1150
295
+ },
296
+ {
297
+ "epoch": 9.33,
298
+ "learning_rate": 8.082857142857144e-06,
299
+ "loss": 0.0111,
300
+ "step": 1175
301
+ },
302
+ {
303
+ "epoch": 9.52,
304
+ "learning_rate": 8.011428571428573e-06,
305
+ "loss": 0.0115,
306
+ "step": 1200
307
+ },
308
+ {
309
+ "epoch": 9.72,
310
+ "learning_rate": 7.94e-06,
311
+ "loss": 0.01,
312
+ "step": 1225
313
+ },
314
+ {
315
+ "epoch": 9.92,
316
+ "learning_rate": 7.86857142857143e-06,
317
+ "loss": 0.013,
318
+ "step": 1250
319
+ },
320
+ {
321
+ "epoch": 10.12,
322
+ "learning_rate": 7.797142857142858e-06,
323
+ "loss": 0.0097,
324
+ "step": 1275
325
+ },
326
+ {
327
+ "epoch": 10.32,
328
+ "learning_rate": 7.725714285714286e-06,
329
+ "loss": 0.0074,
330
+ "step": 1300
331
+ },
332
+ {
333
+ "epoch": 10.52,
334
+ "learning_rate": 7.654285714285715e-06,
335
+ "loss": 0.0075,
336
+ "step": 1325
337
+ },
338
+ {
339
+ "epoch": 10.71,
340
+ "learning_rate": 7.5828571428571444e-06,
341
+ "loss": 0.0094,
342
+ "step": 1350
343
+ },
344
+ {
345
+ "epoch": 10.91,
346
+ "learning_rate": 7.511428571428572e-06,
347
+ "loss": 0.0107,
348
+ "step": 1375
349
+ },
350
+ {
351
+ "epoch": 11.11,
352
+ "learning_rate": 7.440000000000001e-06,
353
+ "loss": 0.0067,
354
+ "step": 1400
355
+ },
356
+ {
357
+ "epoch": 11.31,
358
+ "learning_rate": 7.36857142857143e-06,
359
+ "loss": 0.0058,
360
+ "step": 1425
361
+ },
362
+ {
363
+ "epoch": 11.51,
364
+ "learning_rate": 7.297142857142858e-06,
365
+ "loss": 0.0065,
366
+ "step": 1450
367
+ },
368
+ {
369
+ "epoch": 11.71,
370
+ "learning_rate": 7.225714285714286e-06,
371
+ "loss": 0.0069,
372
+ "step": 1475
373
+ },
374
+ {
375
+ "epoch": 11.9,
376
+ "learning_rate": 7.154285714285715e-06,
377
+ "loss": 0.0057,
378
+ "step": 1500
379
+ },
380
+ {
381
+ "epoch": 12.1,
382
+ "learning_rate": 7.082857142857143e-06,
383
+ "loss": 0.0049,
384
+ "step": 1525
385
+ },
386
+ {
387
+ "epoch": 12.3,
388
+ "learning_rate": 7.011428571428572e-06,
389
+ "loss": 0.0039,
390
+ "step": 1550
391
+ },
392
+ {
393
+ "epoch": 12.5,
394
+ "learning_rate": 6.9400000000000005e-06,
395
+ "loss": 0.0044,
396
+ "step": 1575
397
+ },
398
+ {
399
+ "epoch": 12.7,
400
+ "learning_rate": 6.868571428571429e-06,
401
+ "loss": 0.0043,
402
+ "step": 1600
403
+ },
404
+ {
405
+ "epoch": 12.9,
406
+ "learning_rate": 6.797142857142858e-06,
407
+ "loss": 0.0041,
408
+ "step": 1625
409
+ },
410
+ {
411
+ "epoch": 13.1,
412
+ "learning_rate": 6.725714285714287e-06,
413
+ "loss": 0.0038,
414
+ "step": 1650
415
+ },
416
+ {
417
+ "epoch": 13.29,
418
+ "learning_rate": 6.654285714285716e-06,
419
+ "loss": 0.0027,
420
+ "step": 1675
421
+ },
422
+ {
423
+ "epoch": 13.49,
424
+ "learning_rate": 6.582857142857143e-06,
425
+ "loss": 0.0028,
426
+ "step": 1700
427
+ },
428
+ {
429
+ "epoch": 13.69,
430
+ "learning_rate": 6.511428571428572e-06,
431
+ "loss": 0.0027,
432
+ "step": 1725
433
+ },
434
+ {
435
+ "epoch": 13.89,
436
+ "learning_rate": 6.440000000000001e-06,
437
+ "loss": 0.0027,
438
+ "step": 1750
439
+ },
440
+ {
441
+ "epoch": 14.09,
442
+ "learning_rate": 6.368571428571429e-06,
443
+ "loss": 0.0021,
444
+ "step": 1775
445
+ },
446
+ {
447
+ "epoch": 14.29,
448
+ "learning_rate": 6.297142857142857e-06,
449
+ "loss": 0.0022,
450
+ "step": 1800
451
+ },
452
+ {
453
+ "epoch": 14.48,
454
+ "learning_rate": 6.225714285714286e-06,
455
+ "loss": 0.0033,
456
+ "step": 1825
457
+ },
458
+ {
459
+ "epoch": 14.68,
460
+ "learning_rate": 6.1542857142857145e-06,
461
+ "loss": 0.002,
462
+ "step": 1850
463
+ },
464
+ {
465
+ "epoch": 14.88,
466
+ "learning_rate": 6.0828571428571435e-06,
467
+ "loss": 0.0024,
468
+ "step": 1875
469
+ },
470
+ {
471
+ "epoch": 15.08,
472
+ "learning_rate": 6.011428571428572e-06,
473
+ "loss": 0.0021,
474
+ "step": 1900
475
+ },
476
+ {
477
+ "epoch": 15.28,
478
+ "learning_rate": 5.94e-06,
479
+ "loss": 0.0012,
480
+ "step": 1925
481
+ },
482
+ {
483
+ "epoch": 15.48,
484
+ "learning_rate": 5.868571428571429e-06,
485
+ "loss": 0.0017,
486
+ "step": 1950
487
+ },
488
+ {
489
+ "epoch": 15.67,
490
+ "learning_rate": 5.797142857142858e-06,
491
+ "loss": 0.0016,
492
+ "step": 1975
493
+ },
494
+ {
495
+ "epoch": 15.87,
496
+ "learning_rate": 5.725714285714287e-06,
497
+ "loss": 0.0021,
498
+ "step": 2000
499
+ },
500
+ {
501
+ "epoch": 15.87,
502
+ "eval_loss": 0.7654294371604919,
503
+ "eval_runtime": 475.0372,
504
+ "eval_samples_per_second": 3.97,
505
+ "eval_steps_per_second": 0.124,
506
+ "eval_wer": 52.89689034369886,
507
+ "step": 2000
508
+ },
509
+ {
510
+ "epoch": 16.07,
511
+ "learning_rate": 5.654285714285714e-06,
512
+ "loss": 0.0017,
513
+ "step": 2025
514
+ },
515
+ {
516
+ "epoch": 16.27,
517
+ "learning_rate": 5.582857142857143e-06,
518
+ "loss": 0.0017,
519
+ "step": 2050
520
+ },
521
+ {
522
+ "epoch": 16.47,
523
+ "learning_rate": 5.511428571428572e-06,
524
+ "loss": 0.0013,
525
+ "step": 2075
526
+ },
527
+ {
528
+ "epoch": 16.67,
529
+ "learning_rate": 5.4400000000000004e-06,
530
+ "loss": 0.0012,
531
+ "step": 2100
532
+ },
533
+ {
534
+ "epoch": 16.87,
535
+ "learning_rate": 5.368571428571429e-06,
536
+ "loss": 0.0019,
537
+ "step": 2125
538
+ },
539
+ {
540
+ "epoch": 17.06,
541
+ "learning_rate": 5.297142857142858e-06,
542
+ "loss": 0.0015,
543
+ "step": 2150
544
+ },
545
+ {
546
+ "epoch": 17.26,
547
+ "learning_rate": 5.225714285714286e-06,
548
+ "loss": 0.0016,
549
+ "step": 2175
550
+ },
551
+ {
552
+ "epoch": 17.46,
553
+ "learning_rate": 5.154285714285715e-06,
554
+ "loss": 0.0013,
555
+ "step": 2200
556
+ },
557
+ {
558
+ "epoch": 17.66,
559
+ "learning_rate": 5.082857142857144e-06,
560
+ "loss": 0.0019,
561
+ "step": 2225
562
+ },
563
+ {
564
+ "epoch": 17.86,
565
+ "learning_rate": 5.011428571428571e-06,
566
+ "loss": 0.001,
567
+ "step": 2250
568
+ },
569
+ {
570
+ "epoch": 18.06,
571
+ "learning_rate": 4.94e-06,
572
+ "loss": 0.001,
573
+ "step": 2275
574
+ },
575
+ {
576
+ "epoch": 18.25,
577
+ "learning_rate": 4.868571428571429e-06,
578
+ "loss": 0.0008,
579
+ "step": 2300
580
+ },
581
+ {
582
+ "epoch": 18.45,
583
+ "learning_rate": 4.797142857142857e-06,
584
+ "loss": 0.0009,
585
+ "step": 2325
586
+ },
587
+ {
588
+ "epoch": 18.65,
589
+ "learning_rate": 4.725714285714286e-06,
590
+ "loss": 0.0007,
591
+ "step": 2350
592
+ },
593
+ {
594
+ "epoch": 18.85,
595
+ "learning_rate": 4.6542857142857145e-06,
596
+ "loss": 0.0008,
597
+ "step": 2375
598
+ },
599
+ {
600
+ "epoch": 19.05,
601
+ "learning_rate": 4.5828571428571435e-06,
602
+ "loss": 0.0007,
603
+ "step": 2400
604
+ },
605
+ {
606
+ "epoch": 19.25,
607
+ "learning_rate": 4.511428571428572e-06,
608
+ "loss": 0.0006,
609
+ "step": 2425
610
+ },
611
+ {
612
+ "epoch": 19.44,
613
+ "learning_rate": 4.440000000000001e-06,
614
+ "loss": 0.0006,
615
+ "step": 2450
616
+ },
617
+ {
618
+ "epoch": 19.64,
619
+ "learning_rate": 4.368571428571429e-06,
620
+ "loss": 0.0011,
621
+ "step": 2475
622
+ },
623
+ {
624
+ "epoch": 19.84,
625
+ "learning_rate": 4.297142857142858e-06,
626
+ "loss": 0.0006,
627
+ "step": 2500
628
+ },
629
+ {
630
+ "epoch": 20.04,
631
+ "learning_rate": 4.225714285714286e-06,
632
+ "loss": 0.0006,
633
+ "step": 2525
634
+ },
635
+ {
636
+ "epoch": 20.24,
637
+ "learning_rate": 4.154285714285714e-06,
638
+ "loss": 0.0005,
639
+ "step": 2550
640
+ },
641
+ {
642
+ "epoch": 20.44,
643
+ "learning_rate": 4.082857142857143e-06,
644
+ "loss": 0.0005,
645
+ "step": 2575
646
+ },
647
+ {
648
+ "epoch": 20.63,
649
+ "learning_rate": 4.011428571428571e-06,
650
+ "loss": 0.0005,
651
+ "step": 2600
652
+ },
653
+ {
654
+ "epoch": 20.83,
655
+ "learning_rate": 3.94e-06,
656
+ "loss": 0.0005,
657
+ "step": 2625
658
+ },
659
+ {
660
+ "epoch": 21.03,
661
+ "learning_rate": 3.8685714285714286e-06,
662
+ "loss": 0.0005,
663
+ "step": 2650
664
+ },
665
+ {
666
+ "epoch": 21.23,
667
+ "learning_rate": 3.7971428571428576e-06,
668
+ "loss": 0.0005,
669
+ "step": 2675
670
+ },
671
+ {
672
+ "epoch": 21.43,
673
+ "learning_rate": 3.7257142857142857e-06,
674
+ "loss": 0.0004,
675
+ "step": 2700
676
+ },
677
+ {
678
+ "epoch": 21.63,
679
+ "learning_rate": 3.6542857142857148e-06,
680
+ "loss": 0.0004,
681
+ "step": 2725
682
+ },
683
+ {
684
+ "epoch": 21.83,
685
+ "learning_rate": 3.582857142857143e-06,
686
+ "loss": 0.0004,
687
+ "step": 2750
688
+ },
689
+ {
690
+ "epoch": 22.02,
691
+ "learning_rate": 3.511428571428572e-06,
692
+ "loss": 0.0005,
693
+ "step": 2775
694
+ },
695
+ {
696
+ "epoch": 22.22,
697
+ "learning_rate": 3.44e-06,
698
+ "loss": 0.0004,
699
+ "step": 2800
700
+ },
701
+ {
702
+ "epoch": 22.42,
703
+ "learning_rate": 3.3685714285714287e-06,
704
+ "loss": 0.0004,
705
+ "step": 2825
706
+ },
707
+ {
708
+ "epoch": 22.62,
709
+ "learning_rate": 3.2971428571428577e-06,
710
+ "loss": 0.0004,
711
+ "step": 2850
712
+ },
713
+ {
714
+ "epoch": 22.82,
715
+ "learning_rate": 3.225714285714286e-06,
716
+ "loss": 0.0004,
717
+ "step": 2875
718
+ },
719
+ {
720
+ "epoch": 23.02,
721
+ "learning_rate": 3.154285714285715e-06,
722
+ "loss": 0.0004,
723
+ "step": 2900
724
+ },
725
+ {
726
+ "epoch": 23.21,
727
+ "learning_rate": 3.082857142857143e-06,
728
+ "loss": 0.0004,
729
+ "step": 2925
730
+ },
731
+ {
732
+ "epoch": 23.41,
733
+ "learning_rate": 3.0114285714285716e-06,
734
+ "loss": 0.0004,
735
+ "step": 2950
736
+ },
737
+ {
738
+ "epoch": 23.61,
739
+ "learning_rate": 2.9400000000000002e-06,
740
+ "loss": 0.0004,
741
+ "step": 2975
742
+ },
743
+ {
744
+ "epoch": 23.81,
745
+ "learning_rate": 2.868571428571429e-06,
746
+ "loss": 0.0004,
747
+ "step": 3000
748
+ },
749
+ {
750
+ "epoch": 23.81,
751
+ "eval_loss": 0.8085867762565613,
752
+ "eval_runtime": 479.2865,
753
+ "eval_samples_per_second": 3.935,
754
+ "eval_steps_per_second": 0.123,
755
+ "eval_wer": 53.10420076377523,
756
+ "step": 3000
757
+ },
758
+ {
759
+ "epoch": 24.01,
760
+ "learning_rate": 2.797142857142857e-06,
761
+ "loss": 0.0004,
762
+ "step": 3025
763
+ },
764
+ {
765
+ "epoch": 24.21,
766
+ "learning_rate": 2.725714285714286e-06,
767
+ "loss": 0.0004,
768
+ "step": 3050
769
+ },
770
+ {
771
+ "epoch": 24.4,
772
+ "learning_rate": 2.654285714285714e-06,
773
+ "loss": 0.0004,
774
+ "step": 3075
775
+ },
776
+ {
777
+ "epoch": 24.6,
778
+ "learning_rate": 2.582857142857143e-06,
779
+ "loss": 0.0004,
780
+ "step": 3100
781
+ },
782
+ {
783
+ "epoch": 24.8,
784
+ "learning_rate": 2.5114285714285718e-06,
785
+ "loss": 0.0004,
786
+ "step": 3125
787
+ },
788
+ {
789
+ "epoch": 25.0,
790
+ "learning_rate": 2.4400000000000004e-06,
791
+ "loss": 0.0004,
792
+ "step": 3150
793
+ },
794
+ {
795
+ "epoch": 25.2,
796
+ "learning_rate": 2.3685714285714285e-06,
797
+ "loss": 0.0004,
798
+ "step": 3175
799
+ },
800
+ {
801
+ "epoch": 25.4,
802
+ "learning_rate": 2.297142857142857e-06,
803
+ "loss": 0.0003,
804
+ "step": 3200
805
+ },
806
+ {
807
+ "epoch": 25.6,
808
+ "learning_rate": 2.2257142857142857e-06,
809
+ "loss": 0.0003,
810
+ "step": 3225
811
+ },
812
+ {
813
+ "epoch": 25.79,
814
+ "learning_rate": 2.1542857142857147e-06,
815
+ "loss": 0.0003,
816
+ "step": 3250
817
+ },
818
+ {
819
+ "epoch": 25.99,
820
+ "learning_rate": 2.0828571428571433e-06,
821
+ "loss": 0.0003,
822
+ "step": 3275
823
+ },
824
+ {
825
+ "epoch": 26.19,
826
+ "learning_rate": 2.0114285714285715e-06,
827
+ "loss": 0.0003,
828
+ "step": 3300
829
+ },
830
+ {
831
+ "epoch": 26.39,
832
+ "learning_rate": 1.94e-06,
833
+ "loss": 0.0003,
834
+ "step": 3325
835
+ },
836
+ {
837
+ "epoch": 26.59,
838
+ "learning_rate": 1.8685714285714289e-06,
839
+ "loss": 0.0003,
840
+ "step": 3350
841
+ },
842
+ {
843
+ "epoch": 26.79,
844
+ "learning_rate": 1.7971428571428572e-06,
845
+ "loss": 0.0003,
846
+ "step": 3375
847
+ },
848
+ {
849
+ "epoch": 26.98,
850
+ "learning_rate": 1.7257142857142858e-06,
851
+ "loss": 0.0003,
852
+ "step": 3400
853
+ },
854
+ {
855
+ "epoch": 27.18,
856
+ "learning_rate": 1.6542857142857144e-06,
857
+ "loss": 0.0003,
858
+ "step": 3425
859
+ },
860
+ {
861
+ "epoch": 27.38,
862
+ "learning_rate": 1.582857142857143e-06,
863
+ "loss": 0.0003,
864
+ "step": 3450
865
+ },
866
+ {
867
+ "epoch": 27.58,
868
+ "learning_rate": 1.5114285714285714e-06,
869
+ "loss": 0.0003,
870
+ "step": 3475
871
+ },
872
+ {
873
+ "epoch": 27.78,
874
+ "learning_rate": 1.44e-06,
875
+ "loss": 0.0003,
876
+ "step": 3500
877
+ },
878
+ {
879
+ "epoch": 27.98,
880
+ "learning_rate": 1.3685714285714286e-06,
881
+ "loss": 0.0003,
882
+ "step": 3525
883
+ },
884
+ {
885
+ "epoch": 28.17,
886
+ "learning_rate": 1.2971428571428574e-06,
887
+ "loss": 0.0003,
888
+ "step": 3550
889
+ },
890
+ {
891
+ "epoch": 28.37,
892
+ "learning_rate": 1.2257142857142857e-06,
893
+ "loss": 0.0003,
894
+ "step": 3575
895
+ },
896
+ {
897
+ "epoch": 28.57,
898
+ "learning_rate": 1.1542857142857143e-06,
899
+ "loss": 0.0003,
900
+ "step": 3600
901
+ },
902
+ {
903
+ "epoch": 28.77,
904
+ "learning_rate": 1.082857142857143e-06,
905
+ "loss": 0.0003,
906
+ "step": 3625
907
+ },
908
+ {
909
+ "epoch": 28.97,
910
+ "learning_rate": 1.0114285714285715e-06,
911
+ "loss": 0.0003,
912
+ "step": 3650
913
+ },
914
+ {
915
+ "epoch": 29.17,
916
+ "learning_rate": 9.400000000000001e-07,
917
+ "loss": 0.0003,
918
+ "step": 3675
919
+ },
920
+ {
921
+ "epoch": 29.37,
922
+ "learning_rate": 8.685714285714286e-07,
923
+ "loss": 0.0003,
924
+ "step": 3700
925
+ },
926
+ {
927
+ "epoch": 29.56,
928
+ "learning_rate": 7.971428571428572e-07,
929
+ "loss": 0.0003,
930
+ "step": 3725
931
+ },
932
+ {
933
+ "epoch": 29.76,
934
+ "learning_rate": 7.257142857142857e-07,
935
+ "loss": 0.0003,
936
+ "step": 3750
937
+ },
938
+ {
939
+ "epoch": 29.96,
940
+ "learning_rate": 6.542857142857144e-07,
941
+ "loss": 0.0003,
942
+ "step": 3775
943
+ },
944
+ {
945
+ "epoch": 30.16,
946
+ "learning_rate": 5.82857142857143e-07,
947
+ "loss": 0.0003,
948
+ "step": 3800
949
+ },
950
+ {
951
+ "epoch": 30.36,
952
+ "learning_rate": 5.114285714285714e-07,
953
+ "loss": 0.0003,
954
+ "step": 3825
955
+ },
956
+ {
957
+ "epoch": 30.56,
958
+ "learning_rate": 4.4e-07,
959
+ "loss": 0.0003,
960
+ "step": 3850
961
+ },
962
+ {
963
+ "epoch": 30.75,
964
+ "learning_rate": 3.685714285714286e-07,
965
+ "loss": 0.0003,
966
+ "step": 3875
967
+ },
968
+ {
969
+ "epoch": 30.95,
970
+ "learning_rate": 2.9714285714285715e-07,
971
+ "loss": 0.0003,
972
+ "step": 3900
973
+ },
974
+ {
975
+ "epoch": 31.15,
976
+ "learning_rate": 2.2571428571428574e-07,
977
+ "loss": 0.0003,
978
+ "step": 3925
979
+ },
980
+ {
981
+ "epoch": 31.35,
982
+ "learning_rate": 1.542857142857143e-07,
983
+ "loss": 0.0003,
984
+ "step": 3950
985
+ },
986
+ {
987
+ "epoch": 31.55,
988
+ "learning_rate": 8.285714285714285e-08,
989
+ "loss": 0.0003,
990
+ "step": 3975
991
+ },
992
+ {
993
+ "epoch": 31.75,
994
+ "learning_rate": 1.142857142857143e-08,
995
+ "loss": 0.0003,
996
+ "step": 4000
997
+ },
998
+ {
999
+ "epoch": 31.75,
1000
+ "eval_loss": 0.8255746960639954,
1001
+ "eval_runtime": 479.9003,
1002
+ "eval_samples_per_second": 3.93,
1003
+ "eval_steps_per_second": 0.123,
1004
+ "eval_wer": 53.136933987997814,
1005
+ "step": 4000
1006
+ }
1007
+ ],
1008
+ "logging_steps": 25,
1009
+ "max_steps": 4000,
1010
+ "num_input_tokens_seen": 0,
1011
+ "num_train_epochs": 32,
1012
+ "save_steps": 1000,
1013
+ "total_flos": 3.69389315127509e+19,
1014
+ "train_batch_size": 8,
1015
+ "trial_name": null,
1016
+ "trial_params": null
1017
+ }