Hugging Face's logo

jkazdan
/
collapse_gemma-2-2b_hs2_accumulate_iter2_sftsd2

Safetensors
gemma2
trl
sft
Generated from Trainer
Model card Files Community
jkazdan commited on
Commit
a91fb28
·
verified ·
1 Parent(s): 41326a7

End of training

Browse files
Files changed (4) hide show
  1. README.md +51 -51
  2. model-00001-of-00002.safetensors +1 -1
  3. model-00002-of-00002.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.0982
21
- - Num Input Tokens Seen: 13719864
22
 
23
  ## Model description
24
 
@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
40
  - learning_rate: 8e-06
41
  - train_batch_size: 8
42
  - eval_batch_size: 16
43
- - seed: 0
44
  - gradient_accumulation_steps: 16
45
  - total_train_batch_size: 128
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -53,54 +53,54 @@ The following hyperparameters were used during training:
53
  | Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
54
  |:-------------:|:------:|:----:|:---------------:|:-----------------:|
55
  | No log | 0 | 0 | 1.3956 | 0 |
56
- | 1.5182 | 0.0206 | 5 | 1.3566 | 278016 |
57
- | 1.3436 | 0.0412 | 10 | 1.2405 | 555760 |
58
- | 1.3099 | 0.0618 | 15 | 1.1728 | 837248 |
59
- | 1.2444 | 0.0824 | 20 | 1.1443 | 1121888 |
60
- | 1.1566 | 0.1030 | 25 | 1.1190 | 1405760 |
61
- | 1.1179 | 0.1236 | 30 | 1.1236 | 1685480 |
62
- | 1.0755 | 0.1441 | 35 | 1.1197 | 1969520 |
63
- | 1.0909 | 0.1647 | 40 | 1.1289 | 2256056 |
64
- | 1.0004 | 0.1853 | 45 | 1.1258 | 2535072 |
65
- | 0.9337 | 0.2059 | 50 | 1.1361 | 2820504 |
66
- | 0.9769 | 0.2265 | 55 | 1.1384 | 3097544 |
67
- | 0.9309 | 0.2471 | 60 | 1.1453 | 3381016 |
68
- | 0.8221 | 0.2677 | 65 | 1.1451 | 3662552 |
69
- | 0.8448 | 0.2883 | 70 | 1.1362 | 3944008 |
70
- | 0.8068 | 0.3089 | 75 | 1.1422 | 4228616 |
71
- | 0.7794 | 0.3295 | 80 | 1.1449 | 4518704 |
72
- | 0.839 | 0.3501 | 85 | 1.1377 | 4803488 |
73
- | 0.7914 | 0.3707 | 90 | 1.1424 | 5092912 |
74
- | 0.7824 | 0.3912 | 95 | 1.1396 | 5376328 |
75
- | 0.7763 | 0.4118 | 100 | 1.1373 | 5657216 |
76
- | 0.7058 | 0.4324 | 105 | 1.1450 | 5936696 |
77
- | 0.7919 | 0.4530 | 110 | 1.1338 | 6218640 |
78
- | 0.6291 | 0.4736 | 115 | 1.1381 | 6500728 |
79
- | 0.6368 | 0.4942 | 120 | 1.1359 | 6781720 |
80
- | 0.6676 | 0.5148 | 125 | 1.1343 | 7069904 |
81
- | 0.6567 | 0.5354 | 130 | 1.1299 | 7351616 |
82
- | 0.7838 | 0.5560 | 135 | 1.1330 | 7641760 |
83
- | 0.6401 | 0.5766 | 140 | 1.1291 | 7931072 |
84
- | 0.6275 | 0.5972 | 145 | 1.1238 | 8217432 |
85
- | 0.6238 | 0.6178 | 150 | 1.1258 | 8498184 |
86
- | 0.639 | 0.6384 | 155 | 1.1231 | 8779760 |
87
- | 0.6416 | 0.6589 | 160 | 1.1231 | 9062392 |
88
- | 0.6282 | 0.6795 | 165 | 1.1192 | 9342232 |
89
- | 0.5363 | 0.7001 | 170 | 1.1197 | 9620560 |
90
- | 0.6333 | 0.7207 | 175 | 1.1168 | 9904800 |
91
- | 0.5421 | 0.7413 | 180 | 1.1152 | 10188928 |
92
- | 0.5879 | 0.7619 | 185 | 1.1131 | 10471944 |
93
- | 0.5608 | 0.7825 | 190 | 1.1117 | 10758568 |
94
- | 0.4817 | 0.8031 | 195 | 1.1109 | 11046576 |
95
- | 0.5578 | 0.8237 | 200 | 1.1081 | 11328352 |
96
- | 0.5967 | 0.8443 | 205 | 1.1053 | 11609888 |
97
- | 0.6086 | 0.8649 | 210 | 1.1074 | 11894256 |
98
- | 0.6493 | 0.8855 | 215 | 1.1021 | 12180976 |
99
- | 0.5754 | 0.9060 | 220 | 1.1066 | 12462336 |
100
- | 0.5951 | 0.9266 | 225 | 1.1012 | 12744360 |
101
- | 0.699 | 0.9472 | 230 | 1.1005 | 13035416 |
102
- | 0.5918 | 0.9678 | 235 | 1.1012 | 13324984 |
103
- | 0.6331 | 0.9884 | 240 | 1.0977 | 13606712 |
104
 
105
 
106
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.0989
21
+ - Num Input Tokens Seen: 13720456
22
 
23
  ## Model description
24
 
 
40
  - learning_rate: 8e-06
41
  - train_batch_size: 8
42
  - eval_batch_size: 16
43
+ - seed: 2
44
  - gradient_accumulation_steps: 16
45
  - total_train_batch_size: 128
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 
53
  | Training Loss | Epoch | Step | Validation Loss | Input Tokens Seen |
54
  |:-------------:|:------:|:----:|:---------------:|:-----------------:|
55
  | No log | 0 | 0 | 1.3956 | 0 |
56
+ | 1.5681 | 0.0206 | 5 | 1.3561 | 281208 |
57
+ | 1.3762 | 0.0412 | 10 | 1.2366 | 561408 |
58
+ | 1.2447 | 0.0618 | 15 | 1.1711 | 846520 |
59
+ | 1.2027 | 0.0824 | 20 | 1.1444 | 1123552 |
60
+ | 1.3092 | 0.1030 | 25 | 1.1172 | 1406248 |
61
+ | 1.1897 | 0.1236 | 30 | 1.1186 | 1693528 |
62
+ | 1.0685 | 0.1441 | 35 | 1.1195 | 1979856 |
63
+ | 0.9925 | 0.1647 | 40 | 1.1286 | 2262504 |
64
+ | 1.0026 | 0.1853 | 45 | 1.1277 | 2544760 |
65
+ | 0.9181 | 0.2059 | 50 | 1.1374 | 2825680 |
66
+ | 0.9007 | 0.2265 | 55 | 1.1411 | 3103440 |
67
+ | 0.8626 | 0.2471 | 60 | 1.1421 | 3388104 |
68
+ | 0.8576 | 0.2677 | 65 | 1.1406 | 3668152 |
69
+ | 0.9025 | 0.2883 | 70 | 1.1459 | 3947528 |
70
+ | 0.8566 | 0.3089 | 75 | 1.1449 | 4229392 |
71
+ | 0.8071 | 0.3295 | 80 | 1.1467 | 4514912 |
72
+ | 0.7788 | 0.3501 | 85 | 1.1398 | 4800168 |
73
+ | 0.7999 | 0.3707 | 90 | 1.1427 | 5085472 |
74
+ | 0.7548 | 0.3912 | 95 | 1.1401 | 5370096 |
75
+ | 0.7775 | 0.4118 | 100 | 1.1324 | 5654448 |
76
+ | 0.6659 | 0.4324 | 105 | 1.1390 | 5932488 |
77
+ | 0.7151 | 0.4530 | 110 | 1.1345 | 6217432 |
78
+ | 0.7126 | 0.4736 | 115 | 1.1303 | 6504472 |
79
+ | 0.5812 | 0.4942 | 120 | 1.1395 | 6786136 |
80
+ | 0.7462 | 0.5148 | 125 | 1.1331 | 7075544 |
81
+ | 0.6824 | 0.5354 | 130 | 1.1306 | 7349632 |
82
+ | 0.7777 | 0.5560 | 135 | 1.1333 | 7638056 |
83
+ | 0.614 | 0.5766 | 140 | 1.1285 | 7926232 |
84
+ | 0.6151 | 0.5972 | 145 | 1.1264 | 8206848 |
85
+ | 0.7309 | 0.6178 | 150 | 1.1235 | 8494256 |
86
+ | 0.6219 | 0.6384 | 155 | 1.1226 | 8771192 |
87
+ | 0.6518 | 0.6589 | 160 | 1.1194 | 9060384 |
88
+ | 0.6101 | 0.6795 | 165 | 1.1167 | 9344632 |
89
+ | 0.6374 | 0.7001 | 170 | 1.1139 | 9625824 |
90
+ | 0.6431 | 0.7207 | 175 | 1.1153 | 9909464 |
91
+ | 0.6351 | 0.7413 | 180 | 1.1112 | 10193712 |
92
+ | 0.6205 | 0.7619 | 185 | 1.1099 | 10473824 |
93
+ | 0.5593 | 0.7825 | 190 | 1.1086 | 10757760 |
94
+ | 0.6611 | 0.8031 | 195 | 1.1067 | 11044304 |
95
+ | 0.604 | 0.8237 | 200 | 1.1089 | 11335648 |
96
+ | 0.5985 | 0.8443 | 205 | 1.1045 | 11616672 |
97
+ | 0.6425 | 0.8649 | 210 | 1.1041 | 11904256 |
98
+ | 0.6244 | 0.8855 | 215 | 1.1036 | 12186800 |
99
+ | 0.4801 | 0.9060 | 220 | 1.1015 | 12472520 |
100
+ | 0.5418 | 0.9266 | 225 | 1.1026 | 12757120 |
101
+ | 0.5693 | 0.9472 | 230 | 1.0992 | 13037120 |
102
+ | 0.6361 | 0.9678 | 235 | 1.0997 | 13321752 |
103
+ | 0.5677 | 0.9884 | 240 | 1.0984 | 13608048 |
104
 
105
 
106
  ### Framework versions
model-00001-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:04f56c14f67866a28f54f9df7118df6be9cd1059c7027ee5ac23f0cd3eff6330
3
  size 4988025760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:575c52fb750fc110c4585a4311bedc8a0315c4bf7160901f69698facfa8afbba
3
  size 4988025760
model-00002-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1cd40f3711f782f2dc6359329b8022ef59f26e18efe1354f0448d5f3dbc67c99
3
  size 240691728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e01d149c0092319af907aff5b407291f7779e81f8c0c21525d4e5bee4a1911f9
3
  size 240691728
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d1ed547099b0b09d65378edaa9c24b93ff47a3b0db2e183d5cd3c2adfe3f3e43
3
  size 5560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc2fd30a95bff6973233ba708ca4d1863612d2e2545bbcf081d545be8b1225d7
3
  size 5560