beamaia commited on
Commit
be61463
1 Parent(s): 569829a

Training in progress, epoch 6

Browse files
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85092bad7d086fd774fcdcecddb04e151a94a4251955f258d495af6a2bc541ee
3
  size 26235704
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e552727933b84bb09190a6ecb7d8bbc669472f0efc17be36189f9aa688f2eabc
3
  size 26235704
checkpoint-16932/README.md CHANGED
@@ -4,6 +4,17 @@ library_name: peft
4
  ## Training procedure
5
 
6
 
 
 
 
 
 
 
 
 
 
 
 
7
  The following `bitsandbytes` quantization config was used during training:
8
  - load_in_8bit: False
9
  - load_in_4bit: True
@@ -16,5 +27,6 @@ The following `bitsandbytes` quantization config was used during training:
16
  - bnb_4bit_compute_dtype: bfloat16
17
  ### Framework versions
18
 
 
19
 
20
  - PEFT 0.4.0
 
4
  ## Training procedure
5
 
6
 
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
+
18
  The following `bitsandbytes` quantization config was used during training:
19
  - load_in_8bit: False
20
  - load_in_4bit: True
 
27
  - bnb_4bit_compute_dtype: bfloat16
28
  ### Framework versions
29
 
30
+ - PEFT 0.4.0
31
 
32
  - PEFT 0.4.0
checkpoint-16932/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bec7c5a936a7abc8d5ef91b3a7aafce166627fd034b75f17eb0797c198f405f
3
+ size 26271757
checkpoint-20319/README.md ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: True
16
+ - bnb_4bit_compute_dtype: bfloat16
17
+ ### Framework versions
18
+
19
+
20
+ - PEFT 0.4.0
checkpoint-20319/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Nous-Hermes-Llama2-13b",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-20319/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e552727933b84bb09190a6ecb7d8bbc669472f0efc17be36189f9aa688f2eabc
3
+ size 26235704
checkpoint-20319/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e2d4e01e67fc8e26a7ee8f0b018ca2faedd3e73c68d3e6b507506af0f6036bc
3
+ size 52562821
checkpoint-20319/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03a1388ceb8d090870c58eb7f82e5b9ea9580da874043e08ef7bc1e14bde2e1e
3
+ size 14575
checkpoint-20319/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c834141536809c10d9ca5d68d2e7766490313acfe738bc340dfb661faaf0dbb6
3
+ size 627
checkpoint-20319/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-20319/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-20319/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "legacy": false,
20
+ "model_max_length": 1000000000000000019884624838656,
21
+ "pad_token": null,
22
+ "sp_model_kwargs": {},
23
+ "tokenizer_class": "LlamaTokenizer",
24
+ "unk_token": {
25
+ "__type": "AddedToken",
26
+ "content": "<unk>",
27
+ "lstrip": false,
28
+ "normalized": true,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ }
32
+ }
checkpoint-20319/trainer_state.json ADDED
@@ -0,0 +1,304 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7249947190284729,
3
+ "best_model_checkpoint": "./Zeroshot/01-12-23-NousResearch-Nous-Hermes-Llama2-13b_multilang-dataset-3.0.3-portuguese-2_epochs-10_batch_2/checkpoints/checkpoint-20319",
4
+ "epoch": 6.0,
5
+ "global_step": 20319,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.15,
12
+ "learning_rate": 5.8948611931482576e-05,
13
+ "loss": 1.3183,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.3,
18
+ "learning_rate": 0.00011801535735380982,
19
+ "loss": 0.9343,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.44,
24
+ "learning_rate": 0.00017708210277613703,
25
+ "loss": 0.9005,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.59,
30
+ "learning_rate": 0.00023614884819846425,
31
+ "loss": 0.8859,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.74,
36
+ "learning_rate": 0.0002952155936207915,
37
+ "loss": 0.8707,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.89,
42
+ "learning_rate": 0.00035416420555227406,
43
+ "loss": 0.8523,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 1.0,
48
+ "eval_loss": 0.8459745645523071,
49
+ "eval_runtime": 133.0397,
50
+ "eval_samples_per_second": 11.32,
51
+ "eval_steps_per_second": 1.421,
52
+ "step": 3386
53
+ },
54
+ {
55
+ "epoch": 1.03,
56
+ "learning_rate": 0.00039998666871418663,
57
+ "loss": 0.8395,
58
+ "step": 3500
59
+ },
60
+ {
61
+ "epoch": 1.18,
62
+ "learning_rate": 0.00039960207649127196,
63
+ "loss": 0.8052,
64
+ "step": 4000
65
+ },
66
+ {
67
+ "epoch": 1.33,
68
+ "learning_rate": 0.0003986872720605752,
69
+ "loss": 0.8036,
70
+ "step": 4500
71
+ },
72
+ {
73
+ "epoch": 1.48,
74
+ "learning_rate": 0.00039724468545932536,
75
+ "loss": 0.7902,
76
+ "step": 5000
77
+ },
78
+ {
79
+ "epoch": 1.62,
80
+ "learning_rate": 0.0003952781486968895,
81
+ "loss": 0.7818,
82
+ "step": 5500
83
+ },
84
+ {
85
+ "epoch": 1.77,
86
+ "learning_rate": 0.00039279288557562877,
87
+ "loss": 0.7834,
88
+ "step": 6000
89
+ },
90
+ {
91
+ "epoch": 1.92,
92
+ "learning_rate": 0.0003897954978146717,
93
+ "loss": 0.7686,
94
+ "step": 6500
95
+ },
96
+ {
97
+ "epoch": 2.0,
98
+ "eval_loss": 0.7881951332092285,
99
+ "eval_runtime": 132.7462,
100
+ "eval_samples_per_second": 11.345,
101
+ "eval_steps_per_second": 1.424,
102
+ "step": 6773
103
+ },
104
+ {
105
+ "epoch": 2.07,
106
+ "learning_rate": 0.0003862939475134625,
107
+ "loss": 0.7371,
108
+ "step": 7000
109
+ },
110
+ {
111
+ "epoch": 2.21,
112
+ "learning_rate": 0.00038230601585633047,
113
+ "loss": 0.6893,
114
+ "step": 7500
115
+ },
116
+ {
117
+ "epoch": 2.36,
118
+ "learning_rate": 0.0003778263160223725,
119
+ "loss": 0.6936,
120
+ "step": 8000
121
+ },
122
+ {
123
+ "epoch": 2.51,
124
+ "learning_rate": 0.00037287424793815686,
125
+ "loss": 0.6923,
126
+ "step": 8500
127
+ },
128
+ {
129
+ "epoch": 2.66,
130
+ "learning_rate": 0.0003674629660106825,
131
+ "loss": 0.6862,
132
+ "step": 9000
133
+ },
134
+ {
135
+ "epoch": 2.81,
136
+ "learning_rate": 0.00036160684447787385,
137
+ "loss": 0.6869,
138
+ "step": 9500
139
+ },
140
+ {
141
+ "epoch": 2.95,
142
+ "learning_rate": 0.0003553214392256293,
143
+ "loss": 0.6856,
144
+ "step": 10000
145
+ },
146
+ {
147
+ "epoch": 3.0,
148
+ "eval_loss": 0.7562845945358276,
149
+ "eval_runtime": 132.7178,
150
+ "eval_samples_per_second": 11.347,
151
+ "eval_steps_per_second": 1.424,
152
+ "step": 10159
153
+ },
154
+ {
155
+ "epoch": 3.1,
156
+ "learning_rate": 0.0003486234464660031,
157
+ "loss": 0.6199,
158
+ "step": 10500
159
+ },
160
+ {
161
+ "epoch": 3.25,
162
+ "learning_rate": 0.0003415452256149365,
163
+ "loss": 0.595,
164
+ "step": 11000
165
+ },
166
+ {
167
+ "epoch": 3.4,
168
+ "learning_rate": 0.0003340772155486902,
169
+ "loss": 0.5962,
170
+ "step": 11500
171
+ },
172
+ {
173
+ "epoch": 3.54,
174
+ "learning_rate": 0.0003262690400990177,
175
+ "loss": 0.6019,
176
+ "step": 12000
177
+ },
178
+ {
179
+ "epoch": 3.69,
180
+ "learning_rate": 0.00031811015213363493,
181
+ "loss": 0.6043,
182
+ "step": 12500
183
+ },
184
+ {
185
+ "epoch": 3.84,
186
+ "learning_rate": 0.0003096375227201415,
187
+ "loss": 0.5984,
188
+ "step": 13000
189
+ },
190
+ {
191
+ "epoch": 3.99,
192
+ "learning_rate": 0.0003008736580950757,
193
+ "loss": 0.6011,
194
+ "step": 13500
195
+ },
196
+ {
197
+ "epoch": 4.0,
198
+ "eval_loss": 0.738968551158905,
199
+ "eval_runtime": 132.6823,
200
+ "eval_samples_per_second": 11.35,
201
+ "eval_steps_per_second": 1.424,
202
+ "step": 13546
203
+ },
204
+ {
205
+ "epoch": 4.13,
206
+ "learning_rate": 0.00029184183811650945,
207
+ "loss": 0.5125,
208
+ "step": 14000
209
+ },
210
+ {
211
+ "epoch": 4.28,
212
+ "learning_rate": 0.00028256605442468665,
213
+ "loss": 0.5143,
214
+ "step": 14500
215
+ },
216
+ {
217
+ "epoch": 4.43,
218
+ "learning_rate": 0.00027307094671192217,
219
+ "loss": 0.5171,
220
+ "step": 15000
221
+ },
222
+ {
223
+ "epoch": 4.58,
224
+ "learning_rate": 0.0002633817372710499,
225
+ "loss": 0.5133,
226
+ "step": 15500
227
+ },
228
+ {
229
+ "epoch": 4.72,
230
+ "learning_rate": 0.00025352416399628225,
231
+ "loss": 0.5155,
232
+ "step": 16000
233
+ },
234
+ {
235
+ "epoch": 4.87,
236
+ "learning_rate": 0.0002435244120144539,
237
+ "loss": 0.5217,
238
+ "step": 16500
239
+ },
240
+ {
241
+ "epoch": 5.0,
242
+ "eval_loss": 0.7306132912635803,
243
+ "eval_runtime": 132.674,
244
+ "eval_samples_per_second": 11.351,
245
+ "eval_steps_per_second": 1.425,
246
+ "step": 16932
247
+ },
248
+ {
249
+ "epoch": 5.02,
250
+ "learning_rate": 0.00023342937243316328,
251
+ "loss": 0.5015,
252
+ "step": 17000
253
+ },
254
+ {
255
+ "epoch": 5.17,
256
+ "learning_rate": 0.00022322540906831537,
257
+ "loss": 0.4297,
258
+ "step": 17500
259
+ },
260
+ {
261
+ "epoch": 5.32,
262
+ "learning_rate": 0.00021295975097746452,
263
+ "loss": 0.4336,
264
+ "step": 18000
265
+ },
266
+ {
267
+ "epoch": 5.46,
268
+ "learning_rate": 0.00020265966730197618,
269
+ "loss": 0.4382,
270
+ "step": 18500
271
+ },
272
+ {
273
+ "epoch": 5.61,
274
+ "learning_rate": 0.00019235251862948508,
275
+ "loss": 0.4357,
276
+ "step": 19000
277
+ },
278
+ {
279
+ "epoch": 5.76,
280
+ "learning_rate": 0.0001820862195310506,
281
+ "loss": 0.4411,
282
+ "step": 19500
283
+ },
284
+ {
285
+ "epoch": 5.91,
286
+ "learning_rate": 0.0001718469024885363,
287
+ "loss": 0.4411,
288
+ "step": 20000
289
+ },
290
+ {
291
+ "epoch": 6.0,
292
+ "eval_loss": 0.7249947190284729,
293
+ "eval_runtime": 132.741,
294
+ "eval_samples_per_second": 11.345,
295
+ "eval_steps_per_second": 1.424,
296
+ "step": 20319
297
+ }
298
+ ],
299
+ "max_steps": 33860,
300
+ "num_train_epochs": 10,
301
+ "total_flos": 9.053504138140262e+17,
302
+ "trial_name": null,
303
+ "trial_params": null
304
+ }
checkpoint-20319/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fb2200feb70e8262514df4fd89c2cc8501a63e46178c2943084ed504f58b222
3
+ size 4283
runs/Dec01_03-27-22_baa8ebef2c59/events.out.tfevents.1701401247.baa8ebef2c59.118280.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2055aca2abcc613a80abc12ff6bf7e990cc44ad22f61f4c97202bbb90a53d6ee
3
- size 11308
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8a3faea65740b7e765bbc6ff6a04230df119a05b129a1a67508a19e882bdcd4
3
+ size 12704