Hugging Face's logo

jkazdan
/
llama8b-gsm-real-and-synthetic-sftsd0

Text Generation
Transformers
Safetensors
llama
trl
sft
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
jkazdan commited on
Commit
b934fe4
·
verified ·
1 Parent(s): ca169e1

End of training

Browse files
Files changed (10) hide show
  1. .gitattributes +1 -0
  2. README.md +99 -98
  3. config.json +2 -1
  4. generation_config.json +1 -1
  5. model-00001-of-00004.safetensors +1 -1
  6. model-00002-of-00004.safetensors +1 -1
  7. model-00003-of-00004.safetensors +1 -1
  8. model-00004-of-00004.safetensors +1 -1
  9. tokenizer.json +0 -0
  10. training_args.bin +2 -2
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  license: llama3
3
  base_model: meta-llama/Meta-Llama-3-8B-Instruct
4
  tags:
@@ -17,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.9782
21
- - Num Input Tokens Seen: 2935514
22
 
23
  ## Model description
24
 
@@ -43,7 +44,7 @@ The following hyperparameters were used during training:
43
  - seed: 0
44
  - gradient_accumulation_steps: 16
45
  - total_train_batch_size: 32
46
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: constant_with_warmup
48
  - lr_scheduler_warmup_ratio: 0.05
49
  - num_epochs: 1
@@ -52,103 +53,103 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
54
  |:-------------:|:------:|:----:|:---------------:|:-----------------:|
55
- | No log | 0 | 0 | 1.8070 | 0 |
56
- | 1.6345 | 0.0109 | 5 | 1.7516 | 31948 |
57
- | 1.2886 | 0.0218 | 10 | 1.5325 | 63172 |
58
- | 1.1379 | 0.0327 | 15 | 1.3973 | 94698 |
59
- | 1.0835 | 0.0435 | 20 | 1.3294 | 126426 |
60
- | 0.9535 | 0.0544 | 25 | 1.2683 | 156924 |
61
- | 0.929 | 0.0653 | 30 | 1.2318 | 188244 |
62
- | 0.8979 | 0.0762 | 35 | 1.1586 | 219528 |
63
- | 0.8583 | 0.0871 | 40 | 1.1205 | 250022 |
64
- | 0.8397 | 0.0980 | 45 | 1.0938 | 281636 |
65
- | 0.869 | 0.1089 | 50 | 1.0715 | 314898 |
66
- | 0.737 | 0.1198 | 55 | 1.0599 | 346332 |
67
- | 0.7946 | 0.1306 | 60 | 1.0557 | 378932 |
68
- | 0.7522 | 0.1415 | 65 | 1.0425 | 410158 |
69
- | 0.7502 | 0.1524 | 70 | 1.0387 | 441482 |
70
- | 0.7871 | 0.1633 | 75 | 1.0336 | 474388 |
71
- | 0.8052 | 0.1742 | 80 | 1.0355 | 506222 |
72
- | 0.8551 | 0.1851 | 85 | 1.0503 | 539436 |
73
- | 0.6407 | 0.1960 | 90 | 1.0413 | 571118 |
74
- | 0.7397 | 0.2069 | 95 | 1.0453 | 604064 |
75
- | 0.8441 | 0.2177 | 100 | 1.0376 | 635310 |
76
- | 0.6898 | 0.2286 | 105 | 1.0379 | 668154 |
77
- | 0.9959 | 0.2395 | 110 | 1.0573 | 701682 |
78
- | 0.7643 | 0.2504 | 115 | 1.0520 | 733494 |
79
- | 0.7565 | 0.2613 | 120 | 1.0503 | 766060 |
80
- | 0.8132 | 0.2722 | 125 | 1.0463 | 798646 |
81
- | 0.9402 | 0.2831 | 130 | 1.0615 | 831576 |
82
- | 0.8057 | 0.2940 | 135 | 1.0477 | 863166 |
83
- | 0.8186 | 0.3048 | 140 | 1.0377 | 895748 |
84
- | 0.8453 | 0.3157 | 145 | 1.0569 | 927306 |
85
- | 0.7614 | 0.3266 | 150 | 1.0405 | 960942 |
86
- | 0.8405 | 0.3375 | 155 | 1.0563 | 992550 |
87
- | 0.8718 | 0.3484 | 160 | 1.0334 | 1024826 |
88
- | 0.8158 | 0.3593 | 165 | 1.0444 | 1056818 |
89
- | 0.7939 | 0.3702 | 170 | 1.0406 | 1089076 |
90
- | 0.7566 | 0.3811 | 175 | 1.0430 | 1120326 |
91
- | 0.7303 | 0.3919 | 180 | 1.0253 | 1152172 |
92
- | 0.7143 | 0.4028 | 185 | 1.0272 | 1183232 |
93
- | 0.7855 | 0.4137 | 190 | 1.0257 | 1215904 |
94
- | 0.8264 | 0.4246 | 195 | 1.0280 | 1247254 |
95
- | 0.7625 | 0.4355 | 200 | 1.0259 | 1279592 |
96
- | 0.9388 | 0.4464 | 205 | 1.0191 | 1312810 |
97
- | 0.8368 | 0.4573 | 210 | 1.0153 | 1344552 |
98
- | 0.7935 | 0.4682 | 215 | 1.0130 | 1375678 |
99
- | 0.7358 | 0.4790 | 220 | 1.0164 | 1408250 |
100
- | 0.7998 | 0.4899 | 225 | 1.0129 | 1441310 |
101
- | 0.7808 | 0.5008 | 230 | 1.0117 | 1473814 |
102
- | 0.8598 | 0.5117 | 235 | 1.0120 | 1506286 |
103
- | 0.8092 | 0.5226 | 240 | 1.0102 | 1537676 |
104
- | 0.7033 | 0.5335 | 245 | 1.0077 | 1569230 |
105
- | 0.6772 | 0.5444 | 250 | 1.0074 | 1601132 |
106
- | 0.7523 | 0.5553 | 255 | 1.0180 | 1632784 |
107
- | 0.7491 | 0.5661 | 260 | 1.0068 | 1665664 |
108
- | 0.8321 | 0.5770 | 265 | 1.0003 | 1695848 |
109
- | 0.8089 | 0.5879 | 270 | 1.0056 | 1727680 |
110
- | 0.8189 | 0.5988 | 275 | 1.0007 | 1760026 |
111
- | 0.8385 | 0.6097 | 280 | 1.0073 | 1792096 |
112
- | 0.7341 | 0.6206 | 285 | 1.0001 | 1825692 |
113
- | 0.7625 | 0.6315 | 290 | 0.9999 | 1859346 |
114
- | 0.7663 | 0.6424 | 295 | 1.0077 | 1891664 |
115
- | 0.781 | 0.6532 | 300 | 0.9997 | 1922732 |
116
- | 0.7844 | 0.6641 | 305 | 1.0032 | 1954256 |
117
- | 0.812 | 0.6750 | 310 | 0.9961 | 1986266 |
118
- | 0.6921 | 0.6859 | 315 | 0.9941 | 2017320 |
119
- | 0.7534 | 0.6968 | 320 | 0.9947 | 2048622 |
120
- | 0.825 | 0.7077 | 325 | 0.9975 | 2081244 |
121
- | 0.9064 | 0.7186 | 330 | 0.9981 | 2112508 |
122
- | 0.7432 | 0.7295 | 335 | 0.9945 | 2145440 |
123
- | 0.8073 | 0.7403 | 340 | 0.9932 | 2176330 |
124
- | 0.7577 | 0.7512 | 345 | 0.9922 | 2209184 |
125
- | 0.8518 | 0.7621 | 350 | 0.9906 | 2242062 |
126
- | 0.8262 | 0.7730 | 355 | 0.9923 | 2273718 |
127
- | 0.8452 | 0.7839 | 360 | 0.9887 | 2304698 |
128
- | 0.795 | 0.7948 | 365 | 0.9918 | 2338036 |
129
- | 0.8215 | 0.8057 | 370 | 0.9925 | 2369392 |
130
- | 0.7971 | 0.8165 | 375 | 0.9926 | 2400310 |
131
- | 0.7915 | 0.8274 | 380 | 0.9877 | 2433752 |
132
- | 0.8167 | 0.8383 | 385 | 0.9885 | 2465412 |
133
- | 0.7602 | 0.8492 | 390 | 0.9866 | 2497866 |
134
- | 0.7634 | 0.8601 | 395 | 0.9864 | 2529744 |
135
- | 0.8249 | 0.8710 | 400 | 0.9884 | 2562194 |
136
- | 0.719 | 0.8819 | 405 | 0.9885 | 2594732 |
137
- | 0.8384 | 0.8928 | 410 | 0.9830 | 2625322 |
138
- | 0.6876 | 0.9036 | 415 | 0.9850 | 2656516 |
139
- | 0.7708 | 0.9145 | 420 | 0.9889 | 2688990 |
140
- | 0.7922 | 0.9254 | 425 | 0.9887 | 2720964 |
141
- | 0.7426 | 0.9363 | 430 | 0.9813 | 2750844 |
142
- | 0.7636 | 0.9472 | 435 | 0.9797 | 2782284 |
143
- | 0.7756 | 0.9581 | 440 | 0.9792 | 2815062 |
144
- | 0.7254 | 0.9690 | 445 | 0.9804 | 2845236 |
145
- | 0.8176 | 0.9799 | 450 | 0.9825 | 2878510 |
146
- | 0.7056 | 0.9907 | 455 | 0.9781 | 2910868 |
147
 
148
 
149
  ### Framework versions
150
 
151
- - Transformers 4.44.0
152
- - Pytorch 2.4.0+cu121
153
  - Datasets 2.20.0
154
- - Tokenizers 0.19.1
 
1
  ---
2
+ library_name: transformers
3
  license: llama3
4
  base_model: meta-llama/Meta-Llama-3-8B-Instruct
5
  tags:
 
18
 
19
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.0849
22
+ - Num Input Tokens Seen: 1877420
23
 
24
  ## Model description
25
 
 
44
  - seed: 0
45
  - gradient_accumulation_steps: 16
46
  - total_train_batch_size: 32
47
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: constant_with_warmup
49
  - lr_scheduler_warmup_ratio: 0.05
50
  - num_epochs: 1
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
55
  |:-------------:|:------:|:----:|:---------------:|:-----------------:|
56
+ | No log | 0 | 0 | 1.8595 | 0 |
57
+ | 2.1188 | 0.0109 | 5 | 1.7934 | 20252 |
58
+ | 1.7529 | 0.0218 | 10 | 1.5578 | 40190 |
59
+ | 1.5003 | 0.0327 | 15 | 1.3796 | 60558 |
60
+ | 1.3666 | 0.0435 | 20 | 1.2697 | 80514 |
61
+ | 1.1632 | 0.0544 | 25 | 1.2132 | 100236 |
62
+ | 1.215 | 0.0653 | 30 | 1.1941 | 119622 |
63
+ | 1.2269 | 0.0762 | 35 | 1.1864 | 138484 |
64
+ | 1.1981 | 0.0871 | 40 | 1.1731 | 158578 |
65
+ | 1.125 | 0.0980 | 45 | 1.1734 | 178550 |
66
+ | 1.1639 | 0.1089 | 50 | 1.1657 | 200200 |
67
+ | 1.1696 | 0.1198 | 55 | 1.1645 | 219466 |
68
+ | 1.2649 | 0.1306 | 60 | 1.1570 | 239706 |
69
+ | 1.2061 | 0.1415 | 65 | 1.1551 | 259198 |
70
+ | 1.1787 | 0.1524 | 70 | 1.1528 | 279656 |
71
+ | 1.2122 | 0.1633 | 75 | 1.1465 | 299930 |
72
+ | 1.1786 | 0.1742 | 80 | 1.1467 | 320656 |
73
+ | 1.1947 | 0.1851 | 85 | 1.1454 | 342140 |
74
+ | 1.2227 | 0.1960 | 90 | 1.1418 | 360794 |
75
+ | 1.1515 | 0.2069 | 95 | 1.1423 | 380688 |
76
+ | 1.2093 | 0.2177 | 100 | 1.1362 | 400902 |
77
+ | 1.1598 | 0.2286 | 105 | 1.1337 | 420968 |
78
+ | 1.1775 | 0.2395 | 110 | 1.1316 | 444378 |
79
+ | 1.2074 | 0.2504 | 115 | 1.1301 | 465350 |
80
+ | 1.1737 | 0.2613 | 120 | 1.1305 | 484828 |
81
+ | 1.139 | 0.2722 | 125 | 1.1277 | 506648 |
82
+ | 1.2399 | 0.2831 | 130 | 1.1304 | 528778 |
83
+ | 1.1194 | 0.2940 | 135 | 1.1238 | 549198 |
84
+ | 1.153 | 0.3048 | 140 | 1.1236 | 569690 |
85
+ | 1.207 | 0.3157 | 145 | 1.1232 | 590042 |
86
+ | 1.0488 | 0.3266 | 150 | 1.1236 | 611098 |
87
+ | 1.1494 | 0.3375 | 155 | 1.1202 | 631730 |
88
+ | 1.1719 | 0.3484 | 160 | 1.1183 | 652614 |
89
+ | 1.1237 | 0.3593 | 165 | 1.1177 | 674112 |
90
+ | 1.1495 | 0.3702 | 170 | 1.1181 | 695024 |
91
+ | 1.1714 | 0.3811 | 175 | 1.1162 | 715462 |
92
+ | 1.1136 | 0.3919 | 180 | 1.1163 | 734588 |
93
+ | 1.052 | 0.4028 | 185 | 1.1154 | 753792 |
94
+ | 1.1381 | 0.4137 | 190 | 1.1126 | 774492 |
95
+ | 1.1324 | 0.4246 | 195 | 1.1124 | 794042 |
96
+ | 1.1164 | 0.4355 | 200 | 1.1129 | 813678 |
97
+ | 1.1365 | 0.4464 | 205 | 1.1102 | 835352 |
98
+ | 1.1545 | 0.4573 | 210 | 1.1103 | 854014 |
99
+ | 1.1442 | 0.4682 | 215 | 1.1097 | 873322 |
100
+ | 1.0279 | 0.4790 | 220 | 1.1066 | 894576 |
101
+ | 1.1465 | 0.4899 | 225 | 1.1070 | 915600 |
102
+ | 1.2079 | 0.5008 | 230 | 1.1087 | 935744 |
103
+ | 1.1502 | 0.5117 | 235 | 1.1062 | 956936 |
104
+ | 1.1242 | 0.5226 | 240 | 1.1050 | 977214 |
105
+ | 1.1403 | 0.5335 | 245 | 1.1071 | 996430 |
106
+ | 1.0747 | 0.5444 | 250 | 1.1034 | 1016696 |
107
+ | 1.1064 | 0.5553 | 255 | 1.1034 | 1037988 |
108
+ | 1.0496 | 0.5661 | 260 | 1.1028 | 1058142 |
109
+ | 1.1228 | 0.5770 | 265 | 1.0994 | 1078686 |
110
+ | 1.1253 | 0.5879 | 270 | 1.0994 | 1100626 |
111
+ | 1.1824 | 0.5988 | 275 | 1.0989 | 1121792 |
112
+ | 1.1731 | 0.6097 | 280 | 1.1000 | 1142104 |
113
+ | 1.1854 | 0.6206 | 285 | 1.0987 | 1164394 |
114
+ | 1.1058 | 0.6315 | 290 | 1.0981 | 1185814 |
115
+ | 1.1307 | 0.6424 | 295 | 1.1006 | 1207150 |
116
+ | 1.0745 | 0.6532 | 300 | 1.0995 | 1226836 |
117
+ | 1.0749 | 0.6641 | 305 | 1.0980 | 1248276 |
118
+ | 1.1606 | 0.6750 | 310 | 1.0952 | 1269206 |
119
+ | 1.0947 | 0.6859 | 315 | 1.0951 | 1290778 |
120
+ | 1.1203 | 0.6968 | 320 | 1.0963 | 1311496 |
121
+ | 1.2225 | 0.7077 | 325 | 1.0947 | 1332048 |
122
+ | 1.2869 | 0.7186 | 330 | 1.0957 | 1351234 |
123
+ | 1.1809 | 0.7295 | 335 | 1.0955 | 1372696 |
124
+ | 1.0819 | 0.7403 | 340 | 1.0973 | 1391276 |
125
+ | 1.096 | 0.7512 | 345 | 1.0943 | 1413020 |
126
+ | 1.1196 | 0.7621 | 350 | 1.0925 | 1435058 |
127
+ | 1.0894 | 0.7730 | 355 | 1.0925 | 1455410 |
128
+ | 1.1599 | 0.7839 | 360 | 1.0917 | 1474912 |
129
+ | 1.0866 | 0.7948 | 365 | 1.0919 | 1495480 |
130
+ | 1.2109 | 0.8057 | 370 | 1.0935 | 1515054 |
131
+ | 1.1566 | 0.8165 | 375 | 1.0910 | 1534450 |
132
+ | 1.1502 | 0.8274 | 380 | 1.0885 | 1556162 |
133
+ | 1.1446 | 0.8383 | 385 | 1.0893 | 1577012 |
134
+ | 1.1439 | 0.8492 | 390 | 1.0905 | 1596860 |
135
+ | 1.0844 | 0.8601 | 395 | 1.0904 | 1616948 |
136
+ | 1.1822 | 0.8710 | 400 | 1.0897 | 1636722 |
137
+ | 1.1542 | 0.8819 | 405 | 1.0878 | 1658786 |
138
+ | 1.1622 | 0.8928 | 410 | 1.0861 | 1677850 |
139
+ | 1.0757 | 0.9036 | 415 | 1.0866 | 1697232 |
140
+ | 1.1228 | 0.9145 | 420 | 1.0881 | 1717802 |
141
+ | 1.0552 | 0.9254 | 425 | 1.0860 | 1738272 |
142
+ | 1.0828 | 0.9363 | 430 | 1.0840 | 1757592 |
143
+ | 1.064 | 0.9472 | 435 | 1.0841 | 1777796 |
144
+ | 1.1513 | 0.9581 | 440 | 1.0838 | 1798990 |
145
+ | 1.1968 | 0.9690 | 445 | 1.0843 | 1817942 |
146
+ | 1.111 | 0.9799 | 450 | 1.0840 | 1840536 |
147
+ | 1.1396 | 0.9907 | 455 | 1.0841 | 1861298 |
148
 
149
 
150
  ### Framework versions
151
 
152
+ - Transformers 4.46.0
153
+ - Pytorch 2.4.1.post300
154
  - Datasets 2.20.0
155
+ - Tokenizers 0.20.1
config.json CHANGED
@@ -7,6 +7,7 @@
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 128000,
9
  "eos_token_id": 128009,
 
10
  "hidden_act": "silu",
11
  "hidden_size": 4096,
12
  "initializer_range": 0.02,
@@ -23,7 +24,7 @@
23
  "rope_theta": 500000.0,
24
  "tie_word_embeddings": false,
25
  "torch_dtype": "bfloat16",
26
- "transformers_version": "4.44.0",
27
  "use_cache": true,
28
  "vocab_size": 128257
29
  }
 
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 128000,
9
  "eos_token_id": 128009,
10
+ "head_dim": 128,
11
  "hidden_act": "silu",
12
  "hidden_size": 4096,
13
  "initializer_range": 0.02,
 
24
  "rope_theta": 500000.0,
25
  "tie_word_embeddings": false,
26
  "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.46.0",
28
  "use_cache": true,
29
  "vocab_size": 128257
30
  }
generation_config.json CHANGED
@@ -8,5 +8,5 @@
8
  "max_length": 4096,
9
  "temperature": 0.6,
10
  "top_p": 0.9,
11
- "transformers_version": "4.44.0"
12
  }
 
8
  "max_length": 4096,
9
  "temperature": 0.6,
10
  "top_p": 0.9,
11
+ "transformers_version": "4.46.0"
12
  }
model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b048e5a6f3579d8e0a9e9012c339d7601365e04cc8592c17c3157bee100e4461
3
  size 4976706864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b089e69a4ec5fdd205c481af52df204bffdba40906b3b998aa50647bfe7d4755
3
  size 4976706864
model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5eb9f5d51d7eb55370332f5cd9cc6d552f20bf683bc09f6599e56f0fe58c6294
3
  size 4999802720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaeeaa1fbbd86a56490ba41a8b4ff32cbf80b3b3d40eeeb3b2d42afb0284e823
3
  size 4999802720
model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:77c94fc42b65902028986f99886a50504e90d8de725b2dc62c8ff56813cd527e
3
  size 4915916176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e4e547ebd13d6179d8ea1fc463c3ff344edc58df25a65e4752e1d88be085147
3
  size 4915916176
model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2487f33a4f92fb5ba4f5bb855580aa0d280f7bdb0d97080da36c732a32cd9ddc
3
  size 1168147000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5f61273c2f3e0beb71914c68a99a28b8c27dde5fe5658b627afeb843ddec123
3
  size 1168147000
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:98aa6dc746a0de81afb8deecfd7f6ab1ba32d29038da4ed041d0a66676cc651f
3
- size 5560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:247521a8f322715b24af9b1d516ac844a81a836a880c882f293e43bf5648a224
3
+ size 5624