Hugging Face's logo

CharlesLi
/
OpenELM-1_1B-DPO-full-max-8-reward

Text Generation
Transformers
TensorBoard
Safetensors
openelm
trl
dpo
alignment-handbook
Generated from Trainer
conversational
custom_code
Model card Files Metrics Community
CharlesLi commited on
Commit
3ee5d97
•
1 Parent(s): 53b0e31

Model save

Browse files
Files changed (9) hide show
  1. README.md +37 -37
  2. all_results.json +16 -16
  3. eval_results.json +12 -12
  4. model.safetensors +1 -1
  5. runs/Oct06_22-40-26_xe8545-a100-22/events.out.tfevents.1728260110.xe8545-a100-22.2429082.1 +3 -0
  6. runs/Oct07_13-52-56_xe8545-a100-22/events.out.tfevents.1728302614.xe8545-a100-22.2444076.0 +3 -0
  7. train_results.json +4 -4
  8. trainer_state.json +0 -0
  9. training_args.bin +1 -1
README.md CHANGED
@@ -17,15 +17,15 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model was trained from scratch on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.7073
21
- - Rewards/chosen: -13.0625
22
- - Rewards/rejected: -15.125
23
- - Rewards/accuracies: 0.6289
24
- - Rewards/margins: 2.1094
25
- - Logps/rejected: -1808.0
26
- - Logps/chosen: -1624.0
27
- - Logits/rejected: 1.7188
28
- - Logits/chosen: 0.1670
29
 
30
  ## Model description
31
 
@@ -62,34 +62,34 @@ The following hyperparameters were used during training:
62
 
63
  | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
64
  |:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
65
- | 0.5172 | 0.1047 | 100 | 0.6643 | -1.3984 | -1.6016 | 0.5840 | 0.2109 | -450.0 | -458.0 | -9.875 | -10.3125 |
66
- | 0.4478 | 0.2094 | 200 | 0.7485 | -3.0 | -3.4688 | 0.5996 | 0.4727 | -636.0 | -620.0 | -9.25 | -9.75 |
67
- | 0.4548 | 0.3141 | 300 | 0.8017 | -3.2031 | -3.6406 | 0.5586 | 0.4395 | -652.0 | -640.0 | -8.25 | -9.0625 |
68
- | 0.4474 | 0.4188 | 400 | 0.7341 | -3.1875 | -3.75 | 0.6270 | 0.5625 | -664.0 | -636.0 | -9.125 | -10.0625 |
69
- | 0.398 | 0.5236 | 500 | 0.8245 | -3.7344 | -4.4688 | 0.6367 | 0.7266 | -736.0 | -692.0 | -6.6875 | -8.125 |
70
- | 0.4063 | 0.6283 | 600 | 0.8894 | -5.375 | -6.2188 | 0.6172 | 0.8477 | -912.0 | -856.0 | -8.0625 | -9.1875 |
71
- | 0.3877 | 0.7330 | 700 | 0.8034 | -4.5 | -5.4688 | 0.6406 | 0.9453 | -836.0 | -768.0 | -5.5938 | -7.125 |
72
- | 0.4087 | 0.8377 | 800 | 0.7859 | -4.9688 | -5.75 | 0.6113 | 0.7930 | -864.0 | -816.0 | -4.375 | -5.8438 |
73
- | 0.3409 | 0.9424 | 900 | 0.9967 | -7.4688 | -8.6875 | 0.625 | 1.2031 | -1160.0 | -1064.0 | -2.8594 | -4.3125 |
74
- | 0.095 | 1.0471 | 1000 | 1.2728 | -6.9375 | -8.25 | 0.5996 | 1.3047 | -1112.0 | -1012.0 | -4.5625 | -6.2188 |
75
- | 0.0768 | 1.1518 | 1100 | 1.4990 | -10.375 | -11.9375 | 0.5977 | 1.5625 | -1480.0 | -1352.0 | -0.8711 | -2.4688 |
76
- | 0.1176 | 1.2565 | 1200 | 1.0659 | -6.4688 | -7.5 | 0.6113 | 1.0312 | -1040.0 | -964.0 | -1.0391 | -2.5781 |
77
- | 0.0885 | 1.3613 | 1300 | 1.1804 | -7.375 | -8.75 | 0.6172 | 1.3672 | -1160.0 | -1056.0 | -2.6562 | -4.2812 |
78
- | 0.0942 | 1.4660 | 1400 | 1.1682 | -8.8125 | -10.3125 | 0.6074 | 1.5 | -1320.0 | -1200.0 | 0.0698 | -1.4531 |
79
- | 0.1054 | 1.5707 | 1500 | 1.0602 | -7.1875 | -8.6875 | 0.6348 | 1.5078 | -1160.0 | -1040.0 | -0.0232 | -1.6719 |
80
- | 0.0799 | 1.6754 | 1600 | 1.1262 | -8.4375 | -9.875 | 0.6387 | 1.4453 | -1272.0 | -1160.0 | 0.6484 | -1.0391 |
81
- | 0.0894 | 1.7801 | 1700 | 1.1620 | -8.6875 | -10.1875 | 0.6367 | 1.5234 | -1304.0 | -1184.0 | -0.6094 | -2.125 |
82
- | 0.1011 | 1.8848 | 1800 | 1.0582 | -7.7812 | -9.1875 | 0.6445 | 1.3984 | -1208.0 | -1096.0 | -0.2539 | -1.7891 |
83
- | 0.068 | 1.9895 | 1900 | 1.2470 | -9.25 | -10.8125 | 0.6367 | 1.5625 | -1368.0 | -1248.0 | 0.5078 | -1.0625 |
84
- | 0.0203 | 2.0942 | 2000 | 1.5382 | -11.875 | -13.625 | 0.6211 | 1.7266 | -1648.0 | -1504.0 | 1.3828 | -0.1348 |
85
- | 0.033 | 2.1990 | 2100 | 1.6683 | -12.25 | -14.3125 | 0.6367 | 2.0625 | -1720.0 | -1544.0 | 1.3359 | -0.2432 |
86
- | 0.0123 | 2.3037 | 2200 | 1.6836 | -13.0 | -15.0 | 0.6289 | 2.0312 | -1792.0 | -1616.0 | 1.6641 | 0.1426 |
87
- | 0.0123 | 2.4084 | 2300 | 1.7316 | -13.375 | -15.5625 | 0.6270 | 2.2031 | -1848.0 | -1656.0 | 1.75 | 0.1934 |
88
- | 0.0167 | 2.5131 | 2400 | 1.6171 | -11.8125 | -13.8125 | 0.6309 | 2.0312 | -1672.0 | -1496.0 | 1.4688 | -0.1260 |
89
- | 0.0138 | 2.6178 | 2500 | 1.7572 | -13.4375 | -15.625 | 0.6270 | 2.1406 | -1848.0 | -1664.0 | 1.75 | 0.2041 |
90
- | 0.0097 | 2.7225 | 2600 | 1.6906 | -12.9375 | -15.0625 | 0.6230 | 2.1094 | -1792.0 | -1616.0 | 1.7031 | 0.1445 |
91
- | 0.0149 | 2.8272 | 2700 | 1.7064 | -13.125 | -15.25 | 0.6289 | 2.1094 | -1808.0 | -1632.0 | 1.7344 | 0.1836 |
92
- | 0.0092 | 2.9319 | 2800 | 1.7073 | -13.0625 | -15.125 | 0.6289 | 2.1094 | -1808.0 | -1624.0 | 1.7188 | 0.1670 |
93
 
94
 
95
  ### Framework versions
 
17
 
18
  This model was trained from scratch on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.7740
21
+ - Rewards/chosen: -15.6875
22
+ - Rewards/rejected: -17.875
23
+ - Rewards/accuracies: 0.6172
24
+ - Rewards/margins: 2.2031
25
+ - Logps/rejected: -2080.0
26
+ - Logps/chosen: -1888.0
27
+ - Logits/rejected: 0.8320
28
+ - Logits/chosen: -0.9922
29
 
30
  ## Model description
31
 
 
62
 
63
  | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
64
  |:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
65
+ | 0.5077 | 0.1047 | 100 | 0.6684 | -1.2422 | -1.4922 | 0.6191 | 0.2490 | -438.0 | -442.0 | -10.4375 | -10.8125 |
66
+ | 0.436 | 0.2094 | 200 | 0.7756 | -2.9219 | -3.3281 | 0.6191 | 0.4141 | -620.0 | -608.0 | -9.6875 | -10.1875 |
67
+ | 0.4375 | 0.3141 | 300 | 0.7544 | -4.0 | -4.625 | 0.6426 | 0.6328 | -752.0 | -720.0 | -8.9375 | -9.9375 |
68
+ | 0.4641 | 0.4188 | 400 | 0.7598 | -3.5938 | -4.2188 | 0.6270 | 0.6094 | -708.0 | -680.0 | -9.8125 | -10.6875 |
69
+ | 0.3819 | 0.5236 | 500 | 0.8648 | -5.0938 | -5.8438 | 0.6074 | 0.7383 | -872.0 | -828.0 | -7.8438 | -9.125 |
70
+ | 0.4052 | 0.6283 | 600 | 0.8811 | -5.1875 | -5.9375 | 0.6016 | 0.7461 | -880.0 | -836.0 | -9.3125 | -10.625 |
71
+ | 0.397 | 0.7330 | 700 | 0.7826 | -4.5938 | -5.3438 | 0.6445 | 0.7578 | -824.0 | -780.0 | -7.5 | -9.125 |
72
+ | 0.3853 | 0.8377 | 800 | 0.8263 | -5.8438 | -6.5938 | 0.6328 | 0.7461 | -948.0 | -904.0 | -5.9688 | -7.3125 |
73
+ | 0.3438 | 0.9424 | 900 | 1.0278 | -7.5938 | -8.8125 | 0.6230 | 1.2344 | -1168.0 | -1080.0 | -2.5 | -4.2188 |
74
+ | 0.0879 | 1.0471 | 1000 | 1.2819 | -9.375 | -10.8125 | 0.6055 | 1.4375 | -1368.0 | -1256.0 | -6.625 | -8.5 |
75
+ | 0.0875 | 1.1518 | 1100 | 1.2599 | -10.3125 | -11.75 | 0.6152 | 1.4609 | -1464.0 | -1352.0 | -3.6406 | -5.25 |
76
+ | 0.1119 | 1.2565 | 1200 | 1.0713 | -7.9688 | -9.125 | 0.6230 | 1.1562 | -1200.0 | -1112.0 | -4.375 | -6.2188 |
77
+ | 0.1083 | 1.3613 | 1300 | 1.1731 | -10.1875 | -11.5 | 0.5918 | 1.2969 | -1440.0 | -1336.0 | -3.7188 | -5.4375 |
78
+ | 0.0827 | 1.4660 | 1400 | 1.0477 | -9.25 | -10.5 | 0.6152 | 1.25 | -1336.0 | -1240.0 | -2.6094 | -4.5 |
79
+ | 0.0913 | 1.5707 | 1500 | 1.0557 | -9.25 | -10.625 | 0.6270 | 1.3828 | -1352.0 | -1248.0 | -2.9688 | -4.7812 |
80
+ | 0.0813 | 1.6754 | 1600 | 1.2081 | -11.4375 | -13.0 | 0.6230 | 1.5625 | -1584.0 | -1456.0 | -1.0156 | -2.7812 |
81
+ | 0.0882 | 1.7801 | 1700 | 1.1652 | -11.5625 | -13.0 | 0.6348 | 1.4531 | -1592.0 | -1472.0 | -3.0469 | -4.7812 |
82
+ | 0.0991 | 1.8848 | 1800 | 1.0546 | -9.6875 | -11.0 | 0.6211 | 1.3203 | -1392.0 | -1288.0 | -0.2773 | -2.0469 |
83
+ | 0.0663 | 1.9895 | 1900 | 1.1602 | -11.0625 | -12.625 | 0.6348 | 1.5312 | -1552.0 | -1424.0 | -1.9766 | -3.7344 |
84
+ | 0.0132 | 2.0942 | 2000 | 1.6895 | -15.4375 | -17.5 | 0.6191 | 2.0625 | -2040.0 | -1856.0 | 0.3359 | -1.5391 |
85
+ | 0.0613 | 2.1990 | 2100 | 1.7890 | -15.8125 | -18.0 | 0.6191 | 2.2031 | -2096.0 | -1896.0 | 0.7539 | -1.0625 |
86
+ | 0.0101 | 2.3037 | 2200 | 1.7495 | -16.125 | -18.375 | 0.6211 | 2.2031 | -2128.0 | -1928.0 | 1.25 | -0.4414 |
87
+ | 0.0138 | 2.4084 | 2300 | 1.7596 | -15.625 | -17.75 | 0.6133 | 2.2031 | -2064.0 | -1880.0 | 1.0234 | -0.7891 |
88
+ | 0.0121 | 2.5131 | 2400 | 1.7912 | -15.625 | -17.875 | 0.6152 | 2.2188 | -2080.0 | -1880.0 | 0.6641 | -1.1797 |
89
+ | 0.0107 | 2.6178 | 2500 | 1.7927 | -15.75 | -18.0 | 0.6133 | 2.1875 | -2080.0 | -1896.0 | 0.8281 | -0.9883 |
90
+ | 0.0145 | 2.7225 | 2600 | 1.7578 | -15.5 | -17.625 | 0.6191 | 2.2031 | -2048.0 | -1864.0 | 0.7031 | -1.1328 |
91
+ | 0.0133 | 2.8272 | 2700 | 1.7674 | -15.625 | -17.875 | 0.6152 | 2.2031 | -2080.0 | -1880.0 | 0.8281 | -0.9961 |
92
+ | 0.0114 | 2.9319 | 2800 | 1.7740 | -15.6875 | -17.875 | 0.6172 | 2.2031 | -2080.0 | -1888.0 | 0.8320 | -0.9922 |
93
 
94
 
95
  ### Framework versions
all_results.json CHANGED
@@ -1,22 +1,22 @@
1
  {
2
  "epoch": 3.0,
3
- "eval_logits/chosen": -0.31640625,
4
- "eval_logits/rejected": 1.4375,
5
- "eval_logps/chosen": -1816.0,
6
- "eval_logps/rejected": -2016.0,
7
- "eval_loss": 1.8165313005447388,
8
- "eval_rewards/accuracies": 0.625,
9
- "eval_rewards/chosen": -15.0,
10
- "eval_rewards/margins": 2.3125,
11
- "eval_rewards/rejected": -17.25,
12
- "eval_runtime": 46.8679,
13
  "eval_samples": 2000,
14
- "eval_samples_per_second": 42.673,
15
- "eval_steps_per_second": 0.683,
16
  "total_flos": 0.0,
17
- "train_loss": 0.17899337792687808,
18
- "train_runtime": 12124.1231,
19
  "train_samples": 61119,
20
- "train_samples_per_second": 15.123,
21
- "train_steps_per_second": 0.236
22
  }
 
1
  {
2
  "epoch": 3.0,
3
+ "eval_logits/chosen": 0.1689453125,
4
+ "eval_logits/rejected": 1.71875,
5
+ "eval_logps/chosen": -1624.0,
6
+ "eval_logps/rejected": -1808.0,
7
+ "eval_loss": 1.7063828706741333,
8
+ "eval_rewards/accuracies": 0.626953125,
9
+ "eval_rewards/chosen": -13.0625,
10
+ "eval_rewards/margins": 2.109375,
11
+ "eval_rewards/rejected": -15.1875,
12
+ "eval_runtime": 46.4619,
13
  "eval_samples": 2000,
14
+ "eval_samples_per_second": 43.046,
15
+ "eval_steps_per_second": 0.689,
16
  "total_flos": 0.0,
17
+ "train_loss": 0.17748919836981758,
18
+ "train_runtime": 12763.6539,
19
  "train_samples": 61119,
20
+ "train_samples_per_second": 14.366,
21
+ "train_steps_per_second": 0.224
22
  }
eval_results.json CHANGED
@@ -1,16 +1,16 @@
1
  {
2
  "epoch": 3.0,
3
- "eval_logits/chosen": -0.31640625,
4
- "eval_logits/rejected": 1.4375,
5
- "eval_logps/chosen": -1816.0,
6
- "eval_logps/rejected": -2016.0,
7
- "eval_loss": 1.8165313005447388,
8
- "eval_rewards/accuracies": 0.625,
9
- "eval_rewards/chosen": -15.0,
10
- "eval_rewards/margins": 2.3125,
11
- "eval_rewards/rejected": -17.25,
12
- "eval_runtime": 46.8679,
13
  "eval_samples": 2000,
14
- "eval_samples_per_second": 42.673,
15
- "eval_steps_per_second": 0.683
16
  }
 
1
  {
2
  "epoch": 3.0,
3
+ "eval_logits/chosen": 0.1689453125,
4
+ "eval_logits/rejected": 1.71875,
5
+ "eval_logps/chosen": -1624.0,
6
+ "eval_logps/rejected": -1808.0,
7
+ "eval_loss": 1.7063828706741333,
8
+ "eval_rewards/accuracies": 0.626953125,
9
+ "eval_rewards/chosen": -13.0625,
10
+ "eval_rewards/margins": 2.109375,
11
+ "eval_rewards/rejected": -15.1875,
12
+ "eval_runtime": 46.4619,
13
  "eval_samples": 2000,
14
+ "eval_samples_per_second": 43.046,
15
+ "eval_steps_per_second": 0.689
16
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3cdb594048df8d598dce5133b155664b09b2223836d3dbef235c5a234bdb552
3
  size 2159808696
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a0f201409a90d3e3034b55489863fb5e69c6bd9da0911b50f4d9fa64501fa6b
3
  size 2159808696
runs/Oct06_22-40-26_xe8545-a100-22/events.out.tfevents.1728260110.xe8545-a100-22.2429082.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d8b72f70fb1e319750db4a12a7b44e835bcda789ee4057f35c65f0d399fa8d3
3
+ size 828
runs/Oct07_13-52-56_xe8545-a100-22/events.out.tfevents.1728302614.xe8545-a100-22.2444076.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8711c7f663141750f75cc240f513e9bf87a13d4ea0819473df34ccba97b293e
3
+ size 225699
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "epoch": 3.0,
3
  "total_flos": 0.0,
4
- "train_loss": 0.17899337792687808,
5
- "train_runtime": 12124.1231,
6
  "train_samples": 61119,
7
- "train_samples_per_second": 15.123,
8
- "train_steps_per_second": 0.236
9
  }
 
1
  {
2
  "epoch": 3.0,
3
  "total_flos": 0.0,
4
+ "train_loss": 0.17748919836981758,
5
+ "train_runtime": 12763.6539,
6
  "train_samples": 61119,
7
+ "train_samples_per_second": 14.366,
8
+ "train_steps_per_second": 0.224
9
  }
trainer_state.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 sha256:4588f8ba7518a00b845fa04e6b538633fac32f501b7a5170f937b77a48ccb7c9
3
  size 7672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33f35e4dfa573a1f6ebab349507d215d1349d2c83f9b3577c6c846f3354439f5
3
  size 7672