DewiBrynJones commited on
Commit
77854d0
1 Parent(s): fb70532

End of training

Browse files
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - wer
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # wav2vec2-xlsr-53-ft-btb-cy
17
 
18
- This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 2.2310
21
  - Wer: 0.9240
 
2
  license: apache-2.0
3
  base_model: facebook/wav2vec2-large-xlsr-53
4
  tags:
5
+ - automatic-speech-recognition
6
+ - DewiBrynJones/banc-trawsgrifiadau-bangor-clean
7
  - generated_from_trainer
8
  metrics:
9
  - wer
 
17
 
18
  # wav2vec2-xlsr-53-ft-btb-cy
19
 
20
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN - DEFAULT dataset.
21
  It achieves the following results on the evaluation set:
22
  - Loss: 2.2310
23
  - Wer: 0.9240
all_results.json CHANGED
@@ -1,15 +1,15 @@
1
  {
2
- "epoch": 3.6775106082036775,
3
- "eval_loss": 0.4158582091331482,
4
- "eval_runtime": 162.0151,
5
- "eval_samples": 5656,
6
- "eval_samples_per_second": 34.91,
7
- "eval_steps_per_second": 4.364,
8
- "eval_wer": 0.3171350162892587,
9
- "total_flos": 9.931497341662648e+18,
10
- "train_loss": 1.0819015429570125,
11
- "train_runtime": 7138.5232,
12
- "train_samples": 22621,
13
- "train_samples_per_second": 11.655,
14
- "train_steps_per_second": 0.364
15
  }
 
1
  {
2
+ "epoch": 3.416856492027335,
3
+ "eval_loss": 2.2309672832489014,
4
+ "eval_runtime": 177.4335,
5
+ "eval_samples": 7022,
6
+ "eval_samples_per_second": 39.575,
7
+ "eval_steps_per_second": 4.948,
8
+ "eval_wer": 0.9239559388245924,
9
+ "total_flos": 1.116875973325552e+19,
10
+ "train_loss": 1.37794411722819,
11
+ "train_runtime": 8509.4041,
12
+ "train_samples": 28086,
13
+ "train_samples_per_second": 11.282,
14
+ "train_steps_per_second": 0.705
15
  }
eval_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 3.6775106082036775,
3
- "eval_loss": 0.4158582091331482,
4
- "eval_runtime": 162.0151,
5
- "eval_samples": 5656,
6
- "eval_samples_per_second": 34.91,
7
- "eval_steps_per_second": 4.364,
8
- "eval_wer": 0.3171350162892587
9
  }
 
1
  {
2
+ "epoch": 3.416856492027335,
3
+ "eval_loss": 2.2309672832489014,
4
+ "eval_runtime": 177.4335,
5
+ "eval_samples": 7022,
6
+ "eval_samples_per_second": 39.575,
7
+ "eval_steps_per_second": 4.948,
8
+ "eval_wer": 0.9239559388245924
9
  }
runs/Aug14_15-33-41_a142c9bbb9ba/events.out.tfevents.1723654976.a142c9bbb9ba.30.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed0b0085f51d6227668ab307a090a240b2c41302047ea9b9080c21eb4f2d4579
3
+ size 406
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 3.6775106082036775,
3
- "total_flos": 9.931497341662648e+18,
4
- "train_loss": 1.0819015429570125,
5
- "train_runtime": 7138.5232,
6
- "train_samples": 22621,
7
- "train_samples_per_second": 11.655,
8
- "train_steps_per_second": 0.364
9
  }
 
1
  {
2
+ "epoch": 3.416856492027335,
3
+ "total_flos": 1.116875973325552e+19,
4
+ "train_loss": 1.37794411722819,
5
+ "train_runtime": 8509.4041,
6
+ "train_samples": 28086,
7
+ "train_samples_per_second": 11.282,
8
+ "train_steps_per_second": 0.705
9
  }
trainer_state.json CHANGED
@@ -1,298 +1,395 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 3.6775106082036775,
5
- "eval_steps": 100,
6
- "global_step": 2600,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.14144271570014144,
13
- "eval_loss": 3.3999907970428467,
14
- "eval_runtime": 162.0918,
15
- "eval_samples_per_second": 34.894,
16
- "eval_steps_per_second": 4.362,
17
- "eval_wer": 1.0,
18
- "step": 100
19
- },
20
- {
21
- "epoch": 0.2828854314002829,
22
- "eval_loss": 2.9512617588043213,
23
- "eval_runtime": 159.7591,
24
- "eval_samples_per_second": 35.403,
25
- "eval_steps_per_second": 4.425,
26
  "eval_wer": 1.0,
27
  "step": 200
28
  },
29
  {
30
- "epoch": 0.4243281471004243,
31
- "eval_loss": 1.9453805685043335,
32
- "eval_runtime": 160.8922,
33
- "eval_samples_per_second": 35.154,
34
- "eval_steps_per_second": 4.394,
35
- "eval_wer": 0.9325480252282904,
36
- "step": 300
37
- },
38
- {
39
- "epoch": 0.5657708628005658,
40
- "eval_loss": 1.1412426233291626,
41
- "eval_runtime": 161.9235,
42
- "eval_samples_per_second": 34.93,
43
- "eval_steps_per_second": 4.366,
44
- "eval_wer": 0.773025629503619,
45
  "step": 400
46
  },
47
  {
48
- "epoch": 0.7072135785007072,
49
- "grad_norm": 2.79933500289917,
50
  "learning_rate": 0.00029699999999999996,
51
- "loss": 3.2243,
52
  "step": 500
53
  },
54
  {
55
- "epoch": 0.7072135785007072,
56
- "eval_loss": 0.9250076413154602,
57
- "eval_runtime": 160.568,
58
- "eval_samples_per_second": 35.225,
59
- "eval_steps_per_second": 4.403,
60
- "eval_wer": 0.67890099661376,
61
- "step": 500
62
  },
63
  {
64
- "epoch": 0.8486562942008486,
65
- "eval_loss": 0.8018025159835815,
66
- "eval_runtime": 160.6715,
67
- "eval_samples_per_second": 35.202,
68
- "eval_steps_per_second": 4.4,
69
- "eval_wer": 0.5982892266213028,
70
- "step": 600
71
  },
72
  {
73
- "epoch": 0.9900990099009901,
74
- "eval_loss": 0.7181887030601501,
75
- "eval_runtime": 160.5511,
76
- "eval_samples_per_second": 35.229,
77
- "eval_steps_per_second": 4.404,
78
- "eval_wer": 0.5408675835727239,
79
- "step": 700
80
  },
81
  {
82
- "epoch": 1.1315417256011315,
83
- "eval_loss": 0.6198393106460571,
84
- "eval_runtime": 161.1454,
85
- "eval_samples_per_second": 35.099,
86
- "eval_steps_per_second": 4.387,
87
- "eval_wer": 0.4769944311598273,
88
- "step": 800
89
  },
90
  {
91
- "epoch": 1.272984441301273,
92
- "eval_loss": 0.6102356910705566,
93
- "eval_runtime": 160.7517,
94
- "eval_samples_per_second": 35.185,
95
- "eval_steps_per_second": 4.398,
96
- "eval_wer": 0.47124905714881804,
97
- "step": 900
98
  },
99
  {
100
- "epoch": 1.4144271570014144,
101
- "grad_norm": 1.0946769714355469,
102
- "learning_rate": 0.00022928571428571426,
103
- "loss": 0.7983,
104
- "step": 1000
 
 
105
  },
106
  {
107
- "epoch": 1.4144271570014144,
108
- "eval_loss": 0.5605342984199524,
109
- "eval_runtime": 160.7985,
110
- "eval_samples_per_second": 35.174,
111
- "eval_steps_per_second": 4.397,
112
- "eval_wer": 0.4425863812167996,
113
- "step": 1000
114
  },
115
  {
116
- "epoch": 1.5558698727015559,
117
- "eval_loss": 0.5336170196533203,
118
- "eval_runtime": 160.9878,
119
- "eval_samples_per_second": 35.133,
120
- "eval_steps_per_second": 4.392,
121
- "eval_wer": 0.4162186451830335,
122
- "step": 1100
123
  },
124
  {
125
- "epoch": 1.6973125884016973,
126
- "eval_loss": 0.5258753299713135,
127
- "eval_runtime": 161.623,
128
- "eval_samples_per_second": 34.995,
129
- "eval_steps_per_second": 4.374,
130
- "eval_wer": 0.41161271685577183,
131
- "step": 1200
132
  },
133
  {
134
- "epoch": 1.8387553041018387,
135
- "eval_loss": 0.4960057735443115,
136
- "eval_runtime": 161.2649,
137
- "eval_samples_per_second": 35.073,
138
- "eval_steps_per_second": 4.384,
139
- "eval_wer": 0.3872349986358749,
140
- "step": 1300
141
  },
142
  {
143
- "epoch": 1.9801980198019802,
144
- "eval_loss": 0.48569169640541077,
145
- "eval_runtime": 161.7331,
146
- "eval_samples_per_second": 34.971,
147
- "eval_steps_per_second": 4.371,
148
- "eval_wer": 0.38676959124392163,
149
- "step": 1400
150
  },
151
  {
152
- "epoch": 2.1216407355021216,
153
- "grad_norm": 1.4740066528320312,
154
- "learning_rate": 0.0001582857142857143,
155
- "loss": 0.6274,
156
- "step": 1500
 
 
157
  },
158
  {
159
- "epoch": 2.1216407355021216,
160
- "eval_loss": 0.46894121170043945,
161
- "eval_runtime": 161.6698,
162
- "eval_samples_per_second": 34.985,
163
- "eval_steps_per_second": 4.373,
164
- "eval_wer": 0.36555343358315545,
165
- "step": 1500
166
  },
167
  {
168
- "epoch": 2.263083451202263,
169
- "eval_loss": 0.4679875373840332,
170
- "eval_runtime": 161.7805,
171
- "eval_samples_per_second": 34.961,
172
- "eval_steps_per_second": 4.37,
173
- "eval_wer": 0.3562131886825761,
174
- "step": 1600
175
  },
176
  {
177
- "epoch": 2.4045261669024045,
178
- "eval_loss": 0.4536493718624115,
179
- "eval_runtime": 162.2799,
180
- "eval_samples_per_second": 34.853,
181
- "eval_steps_per_second": 4.357,
182
- "eval_wer": 0.3535330840461556,
183
- "step": 1700
184
  },
185
  {
186
- "epoch": 2.545968882602546,
187
- "eval_loss": 0.44862428307533264,
188
- "eval_runtime": 161.7828,
189
- "eval_samples_per_second": 34.96,
190
- "eval_steps_per_second": 4.37,
191
- "eval_wer": 0.3500826499333986,
192
- "step": 1800
193
  },
194
  {
195
- "epoch": 2.6874115983026874,
196
- "eval_loss": 0.43955981731414795,
197
- "eval_runtime": 162.5762,
198
- "eval_samples_per_second": 34.79,
199
- "eval_steps_per_second": 4.349,
200
- "eval_wer": 0.35051596026383786,
201
- "step": 1900
202
  },
203
  {
204
- "epoch": 2.828854314002829,
205
- "grad_norm": 0.927208662033081,
206
- "learning_rate": 8.728571428571428e-05,
207
- "loss": 0.4939,
208
- "step": 2000
 
 
209
  },
210
  {
211
- "epoch": 2.828854314002829,
212
- "eval_loss": 0.4298574924468994,
213
- "eval_runtime": 162.5099,
214
- "eval_samples_per_second": 34.804,
215
- "eval_steps_per_second": 4.351,
216
- "eval_wer": 0.3340020221148754,
217
- "step": 2000
218
  },
219
  {
220
- "epoch": 2.9702970297029703,
221
- "eval_loss": 0.4291674494743347,
222
- "eval_runtime": 162.0566,
223
- "eval_samples_per_second": 34.901,
224
- "eval_steps_per_second": 4.363,
225
- "eval_wer": 0.33114538364012774,
226
- "step": 2100
227
  },
228
  {
229
- "epoch": 3.1117397454031117,
230
- "eval_loss": 0.42761147022247314,
231
- "eval_runtime": 162.7807,
232
- "eval_samples_per_second": 34.746,
233
- "eval_steps_per_second": 4.343,
234
- "eval_wer": 0.3270690568278474,
235
- "step": 2200
236
  },
237
  {
238
- "epoch": 3.253182461103253,
239
- "eval_loss": 0.4232628345489502,
240
- "eval_runtime": 161.9234,
241
- "eval_samples_per_second": 34.93,
242
- "eval_steps_per_second": 4.366,
243
- "eval_wer": 0.3260419508593988,
244
- "step": 2300
245
  },
246
  {
247
- "epoch": 3.3946251768033946,
248
- "eval_loss": 0.4192351996898651,
249
- "eval_runtime": 162.4093,
250
- "eval_samples_per_second": 34.826,
251
- "eval_steps_per_second": 4.353,
252
- "eval_wer": 0.32230264319301566,
253
- "step": 2400
254
  },
255
  {
256
- "epoch": 3.536067892503536,
257
- "grad_norm": 0.7725916504859924,
258
- "learning_rate": 1.614285714285714e-05,
259
- "loss": 0.4072,
260
- "step": 2500
261
  },
262
  {
263
- "epoch": 3.536067892503536,
264
- "eval_loss": 0.4178767800331116,
265
- "eval_runtime": 162.2884,
266
- "eval_samples_per_second": 34.852,
267
- "eval_steps_per_second": 4.356,
268
- "eval_wer": 0.31946205324902505,
269
- "step": 2500
270
  },
271
  {
272
- "epoch": 3.6775106082036775,
273
- "eval_loss": 0.4158582091331482,
274
- "eval_runtime": 162.2061,
275
- "eval_samples_per_second": 34.869,
276
- "eval_steps_per_second": 4.359,
277
- "eval_wer": 0.3171350162892587,
278
- "step": 2600
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
279
  },
280
  {
281
- "epoch": 3.6775106082036775,
282
- "step": 2600,
283
- "total_flos": 9.931497341662648e+18,
284
- "train_loss": 1.0819015429570125,
285
- "train_runtime": 7138.5232,
286
- "train_samples_per_second": 11.655,
287
- "train_steps_per_second": 0.364
288
  }
289
  ],
290
  "logging_steps": 500,
291
- "max_steps": 2600,
292
  "num_input_tokens_seen": 0,
293
  "num_train_epochs": 4,
294
- "save_steps": 400,
295
- "total_flos": 9.931497341662648e+18,
 
 
 
 
 
 
 
 
 
 
 
 
296
  "train_batch_size": 16,
297
  "trial_name": null,
298
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 3.416856492027335,
5
+ "eval_steps": 200,
6
+ "global_step": 6000,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.11389521640091116,
13
+ "eval_loss": 3.107494354248047,
14
+ "eval_runtime": 174.2934,
15
+ "eval_samples_per_second": 40.288,
16
+ "eval_steps_per_second": 5.037,
 
 
 
 
 
 
 
 
 
17
  "eval_wer": 1.0,
18
  "step": 200
19
  },
20
  {
21
+ "epoch": 0.22779043280182232,
22
+ "eval_loss": 1.6248737573623657,
23
+ "eval_runtime": 174.1392,
24
+ "eval_samples_per_second": 40.324,
25
+ "eval_steps_per_second": 5.042,
26
+ "eval_wer": 0.8958131253404908,
 
 
 
 
 
 
 
 
 
27
  "step": 400
28
  },
29
  {
30
+ "epoch": 0.2847380410022779,
31
+ "grad_norm": 3.3784055709838867,
32
  "learning_rate": 0.00029699999999999996,
33
+ "loss": 3.6043,
34
  "step": 500
35
  },
36
  {
37
+ "epoch": 0.3416856492027335,
38
+ "eval_loss": 1.1246436834335327,
39
+ "eval_runtime": 174.2087,
40
+ "eval_samples_per_second": 40.308,
41
+ "eval_steps_per_second": 5.04,
42
+ "eval_wer": 0.8675772997249498,
43
+ "step": 600
44
  },
45
  {
46
+ "epoch": 0.45558086560364464,
47
+ "eval_loss": 0.9142205715179443,
48
+ "eval_runtime": 174.1428,
49
+ "eval_samples_per_second": 40.323,
50
+ "eval_steps_per_second": 5.042,
51
+ "eval_wer": 0.7093902390411778,
52
+ "step": 800
53
  },
54
  {
55
+ "epoch": 0.5694760820045558,
56
+ "grad_norm": 2.1683263778686523,
57
+ "learning_rate": 0.0002730545454545454,
58
+ "loss": 1.0665,
59
+ "step": 1000
 
 
60
  },
61
  {
62
+ "epoch": 0.5694760820045558,
63
+ "eval_loss": 0.7569850087165833,
64
+ "eval_runtime": 173.8805,
65
+ "eval_samples_per_second": 40.384,
66
+ "eval_steps_per_second": 5.049,
67
+ "eval_wer": 0.5934306860309067,
68
+ "step": 1000
69
  },
70
  {
71
+ "epoch": 0.683371298405467,
72
+ "eval_loss": 0.6979650259017944,
73
+ "eval_runtime": 174.537,
74
+ "eval_samples_per_second": 40.232,
75
+ "eval_steps_per_second": 5.03,
76
+ "eval_wer": 0.5284816433914881,
77
+ "step": 1200
78
  },
79
  {
80
+ "epoch": 0.7972665148063781,
81
+ "eval_loss": 0.6941590905189514,
82
+ "eval_runtime": 174.9006,
83
+ "eval_samples_per_second": 40.149,
84
+ "eval_steps_per_second": 5.02,
85
+ "eval_wer": 0.5388990021127041,
86
+ "step": 1400
87
  },
88
  {
89
+ "epoch": 0.8542141230068337,
90
+ "grad_norm": 2.244067907333374,
91
+ "learning_rate": 0.0002459454545454545,
92
+ "loss": 0.8811,
93
+ "step": 1500
 
 
94
  },
95
  {
96
+ "epoch": 0.9111617312072893,
97
+ "eval_loss": 0.6152143478393555,
98
+ "eval_runtime": 175.8479,
99
+ "eval_samples_per_second": 39.932,
100
+ "eval_steps_per_second": 4.993,
101
+ "eval_wer": 0.4861611235865478,
102
+ "step": 1600
103
  },
104
  {
105
+ "epoch": 1.0250569476082005,
106
+ "eval_loss": 0.5941163301467896,
107
+ "eval_runtime": 174.9993,
108
+ "eval_samples_per_second": 40.126,
109
+ "eval_steps_per_second": 5.017,
110
+ "eval_wer": 0.4534607156619142,
111
+ "step": 1800
112
  },
113
  {
114
+ "epoch": 1.1389521640091116,
115
+ "grad_norm": 2.146209478378296,
116
+ "learning_rate": 0.00021878181818181817,
117
+ "loss": 0.7654,
118
+ "step": 2000
 
 
119
  },
120
  {
121
+ "epoch": 1.1389521640091116,
122
+ "eval_loss": 0.5642105340957642,
123
+ "eval_runtime": 174.8742,
124
+ "eval_samples_per_second": 40.155,
125
+ "eval_steps_per_second": 5.021,
126
+ "eval_wer": 0.43286517227175486,
127
+ "step": 2000
128
  },
129
  {
130
+ "epoch": 1.2528473804100229,
131
+ "eval_loss": 0.5727251768112183,
132
+ "eval_runtime": 175.3644,
133
+ "eval_samples_per_second": 40.042,
134
+ "eval_steps_per_second": 5.007,
135
+ "eval_wer": 0.43133711582667855,
136
+ "step": 2200
137
  },
138
  {
139
+ "epoch": 1.366742596810934,
140
+ "eval_loss": 0.5466997027397156,
141
+ "eval_runtime": 175.0585,
142
+ "eval_samples_per_second": 40.112,
143
+ "eval_steps_per_second": 5.015,
144
+ "eval_wer": 0.4316825894577393,
145
+ "step": 2400
146
  },
147
  {
148
+ "epoch": 1.4236902050113895,
149
+ "grad_norm": 1.5887110233306885,
150
+ "learning_rate": 0.00019167272727272725,
151
+ "loss": 0.6896,
152
+ "step": 2500
 
 
153
  },
154
  {
155
+ "epoch": 1.4806378132118452,
156
+ "eval_loss": 0.5658189654350281,
157
+ "eval_runtime": 178.0734,
158
+ "eval_samples_per_second": 39.433,
159
+ "eval_steps_per_second": 4.931,
160
+ "eval_wer": 0.43980121978766656,
161
+ "step": 2600
162
  },
163
  {
164
+ "epoch": 1.5945330296127562,
165
+ "eval_loss": 0.6008076071739197,
166
+ "eval_runtime": 175.5831,
167
+ "eval_samples_per_second": 39.992,
168
+ "eval_steps_per_second": 5.0,
169
+ "eval_wer": 0.4445182635963805,
170
+ "step": 2800
171
  },
172
  {
173
+ "epoch": 1.7084282460136673,
174
+ "grad_norm": 3.4384372234344482,
175
+ "learning_rate": 0.0001645090909090909,
176
+ "loss": 0.7558,
177
+ "step": 3000
 
 
178
  },
179
  {
180
+ "epoch": 1.7084282460136673,
181
+ "eval_loss": 0.7089642286300659,
182
+ "eval_runtime": 175.5809,
183
+ "eval_samples_per_second": 39.993,
184
+ "eval_steps_per_second": 5.001,
185
+ "eval_wer": 0.5020927729573872,
186
+ "step": 3000
187
  },
188
  {
189
+ "epoch": 1.8223234624145785,
190
+ "eval_loss": 0.8049871921539307,
191
+ "eval_runtime": 175.5153,
192
+ "eval_samples_per_second": 40.008,
193
+ "eval_steps_per_second": 5.002,
194
+ "eval_wer": 0.5291327283115641,
195
+ "step": 3200
196
  },
197
  {
198
+ "epoch": 1.9362186788154898,
199
+ "eval_loss": 0.824817955493927,
200
+ "eval_runtime": 175.9347,
201
+ "eval_samples_per_second": 39.913,
202
+ "eval_steps_per_second": 4.99,
203
+ "eval_wer": 0.5275382346297453,
204
+ "step": 3400
205
  },
206
  {
207
+ "epoch": 1.9931662870159452,
208
+ "grad_norm": 3.6540627479553223,
209
+ "learning_rate": 0.00013723636363636361,
210
+ "loss": 0.9823,
211
+ "step": 3500
 
 
212
  },
213
  {
214
+ "epoch": 2.050113895216401,
215
+ "eval_loss": 0.7944601774215698,
216
+ "eval_runtime": 176.7964,
217
+ "eval_samples_per_second": 39.718,
218
+ "eval_steps_per_second": 4.966,
219
+ "eval_wer": 0.5072748774232982,
220
+ "step": 3600
221
  },
222
  {
223
+ "epoch": 2.164009111617312,
224
+ "eval_loss": 0.7690120935440063,
225
+ "eval_runtime": 177.7491,
226
+ "eval_samples_per_second": 39.505,
227
+ "eval_steps_per_second": 4.94,
228
+ "eval_wer": 0.4982128383316281,
229
+ "step": 3800
230
  },
231
  {
232
+ "epoch": 2.277904328018223,
233
+ "grad_norm": 12.11134147644043,
234
+ "learning_rate": 0.00011001818181818181,
235
+ "loss": 0.9181,
236
+ "step": 4000
237
  },
238
  {
239
+ "epoch": 2.277904328018223,
240
+ "eval_loss": 0.8668798804283142,
241
+ "eval_runtime": 176.748,
242
+ "eval_samples_per_second": 39.729,
243
+ "eval_steps_per_second": 4.968,
244
+ "eval_wer": 0.6929802415657927,
245
+ "step": 4000
246
  },
247
  {
248
+ "epoch": 2.3917995444191344,
249
+ "eval_loss": 0.88874351978302,
250
+ "eval_runtime": 177.153,
251
+ "eval_samples_per_second": 39.638,
252
+ "eval_steps_per_second": 4.956,
253
+ "eval_wer": 0.6311271741585724,
254
+ "step": 4200
255
+ },
256
+ {
257
+ "epoch": 2.5056947608200457,
258
+ "eval_loss": 1.150854468345642,
259
+ "eval_runtime": 176.8956,
260
+ "eval_samples_per_second": 39.696,
261
+ "eval_steps_per_second": 4.963,
262
+ "eval_wer": 0.8436200321556225,
263
+ "step": 4400
264
+ },
265
+ {
266
+ "epoch": 2.562642369020501,
267
+ "grad_norm": 6.307694911956787,
268
+ "learning_rate": 8.285454545454545e-05,
269
+ "loss": 1.0805,
270
+ "step": 4500
271
+ },
272
+ {
273
+ "epoch": 2.619589977220957,
274
+ "eval_loss": 1.2166552543640137,
275
+ "eval_runtime": 176.6784,
276
+ "eval_samples_per_second": 39.745,
277
+ "eval_steps_per_second": 4.969,
278
+ "eval_wer": 0.9458935143969492,
279
+ "step": 4600
280
+ },
281
+ {
282
+ "epoch": 2.733485193621868,
283
+ "eval_loss": 1.2323366403579712,
284
+ "eval_runtime": 177.2283,
285
+ "eval_samples_per_second": 39.621,
286
+ "eval_steps_per_second": 4.954,
287
+ "eval_wer": 0.9307059620776252,
288
+ "step": 4800
289
+ },
290
+ {
291
+ "epoch": 2.847380410022779,
292
+ "grad_norm": 0.8214463591575623,
293
+ "learning_rate": 5.569090909090908e-05,
294
+ "loss": 1.3977,
295
+ "step": 5000
296
+ },
297
+ {
298
+ "epoch": 2.847380410022779,
299
+ "eval_loss": 1.65969979763031,
300
+ "eval_runtime": 176.9601,
301
+ "eval_samples_per_second": 39.681,
302
+ "eval_steps_per_second": 4.962,
303
+ "eval_wer": 0.9527896995708155,
304
+ "step": 5000
305
+ },
306
+ {
307
+ "epoch": 2.9612756264236904,
308
+ "eval_loss": 2.022368907928467,
309
+ "eval_runtime": 176.5944,
310
+ "eval_samples_per_second": 39.763,
311
+ "eval_steps_per_second": 4.972,
312
+ "eval_wer": 0.9162359319151198,
313
+ "step": 5200
314
+ },
315
+ {
316
+ "epoch": 3.075170842824601,
317
+ "eval_loss": 2.105426788330078,
318
+ "eval_runtime": 176.3683,
319
+ "eval_samples_per_second": 39.814,
320
+ "eval_steps_per_second": 4.978,
321
+ "eval_wer": 0.9103761676344357,
322
+ "step": 5400
323
+ },
324
+ {
325
+ "epoch": 3.132118451025057,
326
+ "grad_norm": 0.6184139251708984,
327
+ "learning_rate": 2.8472727272727272e-05,
328
+ "loss": 2.1028,
329
+ "step": 5500
330
+ },
331
+ {
332
+ "epoch": 3.1890660592255125,
333
+ "eval_loss": 2.248642683029175,
334
+ "eval_runtime": 175.7372,
335
+ "eval_samples_per_second": 39.957,
336
+ "eval_steps_per_second": 4.996,
337
+ "eval_wer": 0.9280750475026243,
338
+ "step": 5600
339
+ },
340
+ {
341
+ "epoch": 3.3029612756264237,
342
+ "eval_loss": 2.2446603775024414,
343
+ "eval_runtime": 176.6127,
344
+ "eval_samples_per_second": 39.759,
345
+ "eval_steps_per_second": 4.971,
346
+ "eval_wer": 0.9215774857492127,
347
+ "step": 5800
348
+ },
349
+ {
350
+ "epoch": 3.416856492027335,
351
+ "grad_norm": 0.6574464440345764,
352
+ "learning_rate": 1.3636363636363634e-06,
353
+ "loss": 2.2911,
354
+ "step": 6000
355
+ },
356
+ {
357
+ "epoch": 3.416856492027335,
358
+ "eval_loss": 2.2309672832489014,
359
+ "eval_runtime": 176.1017,
360
+ "eval_samples_per_second": 39.875,
361
+ "eval_steps_per_second": 4.986,
362
+ "eval_wer": 0.9239559388245924,
363
+ "step": 6000
364
  },
365
  {
366
+ "epoch": 3.416856492027335,
367
+ "step": 6000,
368
+ "total_flos": 1.116875973325552e+19,
369
+ "train_loss": 1.37794411722819,
370
+ "train_runtime": 8509.4041,
371
+ "train_samples_per_second": 11.282,
372
+ "train_steps_per_second": 0.705
373
  }
374
  ],
375
  "logging_steps": 500,
376
+ "max_steps": 6000,
377
  "num_input_tokens_seen": 0,
378
  "num_train_epochs": 4,
379
+ "save_steps": 500,
380
+ "stateful_callbacks": {
381
+ "TrainerControl": {
382
+ "args": {
383
+ "should_epoch_stop": false,
384
+ "should_evaluate": false,
385
+ "should_log": false,
386
+ "should_save": true,
387
+ "should_training_stop": true
388
+ },
389
+ "attributes": {}
390
+ }
391
+ },
392
+ "total_flos": 1.116875973325552e+19,
393
  "train_batch_size": 16,
394
  "trial_name": null,
395
  "trial_params": null