Training in progress, epoch 0
Browse files- README.md +15 -13
- adapter_config.json +1 -1
- adapter_model.safetensors +2 -2
- all_results.json +17 -17
- eval_results.json +13 -13
- runs/Dec09_14-07-41_uclaml03.cs.ucla.edu/events.out.tfevents.1702159748.uclaml03.cs.ucla.edu.2757530.0 +3 -0
- runs/Dec09_15-37-03_uclaml03.cs.ucla.edu/events.out.tfevents.1702165262.uclaml03.cs.ucla.edu.2770644.0 +3 -0
- runs/Dec09_15-53-30_uclaml03.cs.ucla.edu/events.out.tfevents.1702166063.uclaml03.cs.ucla.edu.2772424.0 +3 -0
- runs/Dec09_15-56-00_uclaml03.cs.ucla.edu/events.out.tfevents.1702166213.uclaml03.cs.ucla.edu.2772782.0 +3 -0
- runs/Dec09_15-58-05_uclaml03.cs.ucla.edu/events.out.tfevents.1702166345.uclaml03.cs.ucla.edu.2773187.0 +3 -0
- runs/Dec13_15-13-52_uclaml03.cs.ucla.edu/events.out.tfevents.1702509291.uclaml03.cs.ucla.edu.3092760.0 +3 -0
- runs/Dec13_15-25-44_uclaml03.cs.ucla.edu/events.out.tfevents.1702510148.uclaml03.cs.ucla.edu.3094406.0 +3 -0
- train_results.json +5 -5
- trainer_state.json +0 -0
- training_args.bin +2 -2
README.md
CHANGED
@@ -15,15 +15,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 0.
|
19 |
-
- Rewards/chosen: 0.
|
20 |
-
- Rewards/rejected: 0.
|
21 |
-
- Rewards/accuracies: 0.
|
22 |
-
- Rewards/margins: 0.
|
23 |
-
- Logps/rejected: -
|
24 |
-
- Logps/chosen: -
|
25 |
-
- Logits/rejected: -
|
26 |
-
- Logits/chosen: -2.
|
27 |
|
28 |
## Model description
|
29 |
|
@@ -43,24 +43,26 @@ More information needed
|
|
43 |
|
44 |
The following hyperparameters were used during training:
|
45 |
- learning_rate: 5e-07
|
46 |
-
- train_batch_size:
|
47 |
- eval_batch_size: 4
|
48 |
- seed: 42
|
49 |
- distributed_type: multi-GPU
|
50 |
- num_devices: 4
|
51 |
-
- gradient_accumulation_steps:
|
52 |
- total_train_batch_size: 64
|
53 |
- total_eval_batch_size: 16
|
54 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
55 |
- lr_scheduler_type: linear
|
56 |
- lr_scheduler_warmup_ratio: 0.1
|
57 |
-
- num_epochs:
|
58 |
|
59 |
### Training results
|
60 |
|
61 |
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
62 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
63 |
-
| 0.
|
|
|
|
|
64 |
|
65 |
|
66 |
### Framework versions
|
|
|
15 |
|
16 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.5266
|
19 |
+
- Rewards/chosen: -0.1470
|
20 |
+
- Rewards/rejected: -0.8985
|
21 |
+
- Rewards/accuracies: 0.7460
|
22 |
+
- Rewards/margins: 0.7516
|
23 |
+
- Logps/rejected: -228.2694
|
24 |
+
- Logps/chosen: -266.1304
|
25 |
+
- Logits/rejected: -1.9412
|
26 |
+
- Logits/chosen: -2.0659
|
27 |
|
28 |
## Model description
|
29 |
|
|
|
43 |
|
44 |
The following hyperparameters were used during training:
|
45 |
- learning_rate: 5e-07
|
46 |
+
- train_batch_size: 2
|
47 |
- eval_batch_size: 4
|
48 |
- seed: 42
|
49 |
- distributed_type: multi-GPU
|
50 |
- num_devices: 4
|
51 |
+
- gradient_accumulation_steps: 8
|
52 |
- total_train_batch_size: 64
|
53 |
- total_eval_batch_size: 16
|
54 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
55 |
- lr_scheduler_type: linear
|
56 |
- lr_scheduler_warmup_ratio: 0.1
|
57 |
+
- num_epochs: 3
|
58 |
|
59 |
### Training results
|
60 |
|
61 |
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
62 |
|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
63 |
+
| 0.5501 | 1.0 | 968 | 0.5560 | -0.1093 | -0.6400 | 0.7200 | 0.5307 | -225.6847 | -265.7538 | -1.9901 | -2.1115 |
|
64 |
+
| 0.5412 | 2.0 | 1936 | 0.5318 | -0.1497 | -0.8640 | 0.7420 | 0.7143 | -227.9245 | -266.1583 | -1.9509 | -2.0748 |
|
65 |
+
| 0.5454 | 3.0 | 2904 | 0.5266 | -0.1470 | -0.8985 | 0.7460 | 0.7516 | -228.2694 | -266.1304 | -1.9412 | -2.0659 |
|
66 |
|
67 |
|
68 |
### Framework versions
|
adapter_config.json
CHANGED
@@ -16,9 +16,9 @@
|
|
16 |
"rank_pattern": {},
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
|
|
19 |
"v_proj",
|
20 |
"q_proj",
|
21 |
-
"o_proj",
|
22 |
"k_proj"
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
|
|
16 |
"rank_pattern": {},
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
19 |
+
"o_proj",
|
20 |
"v_proj",
|
21 |
"q_proj",
|
|
|
22 |
"k_proj"
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9c393ebd4252a03b80a5249d58bc2098224b1be8de7d9bef2676336f9dd92052
|
3 |
+
size 218138576
|
all_results.json
CHANGED
@@ -1,21 +1,21 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_logits/chosen": -2.
|
4 |
-
"eval_logits/rejected": -
|
5 |
-
"eval_logps/chosen": -
|
6 |
-
"eval_logps/rejected": -
|
7 |
-
"eval_loss": 0.
|
8 |
-
"eval_rewards/accuracies": 0.
|
9 |
-
"eval_rewards/chosen": 0.
|
10 |
-
"eval_rewards/margins": 0.
|
11 |
-
"eval_rewards/rejected": 0.
|
12 |
-
"eval_runtime":
|
13 |
"eval_samples": 2000,
|
14 |
-
"eval_samples_per_second": 4.
|
15 |
-
"eval_steps_per_second": 0.
|
16 |
-
"train_loss": 0.
|
17 |
-
"train_runtime":
|
18 |
"train_samples": 61966,
|
19 |
-
"train_samples_per_second": 2.
|
20 |
-
"train_steps_per_second": 0.
|
21 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_logits/chosen": -2.0658671855926514,
|
4 |
+
"eval_logits/rejected": -1.9411793947219849,
|
5 |
+
"eval_logps/chosen": -266.13037109375,
|
6 |
+
"eval_logps/rejected": -228.2694091796875,
|
7 |
+
"eval_loss": 0.5265706777572632,
|
8 |
+
"eval_rewards/accuracies": 0.7459999918937683,
|
9 |
+
"eval_rewards/chosen": -0.1469534933567047,
|
10 |
+
"eval_rewards/margins": 0.7515553832054138,
|
11 |
+
"eval_rewards/rejected": -0.8985088467597961,
|
12 |
+
"eval_runtime": 453.4186,
|
13 |
"eval_samples": 2000,
|
14 |
+
"eval_samples_per_second": 4.411,
|
15 |
+
"eval_steps_per_second": 0.276,
|
16 |
+
"train_loss": 0.5642705829018732,
|
17 |
+
"train_runtime": 71197.0849,
|
18 |
"train_samples": 61966,
|
19 |
+
"train_samples_per_second": 2.611,
|
20 |
+
"train_steps_per_second": 0.041
|
21 |
}
|
eval_results.json
CHANGED
@@ -1,16 +1,16 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_logits/chosen": -2.
|
4 |
-
"eval_logits/rejected": -
|
5 |
-
"eval_logps/chosen": -
|
6 |
-
"eval_logps/rejected": -
|
7 |
-
"eval_loss": 0.
|
8 |
-
"eval_rewards/accuracies": 0.
|
9 |
-
"eval_rewards/chosen": 0.
|
10 |
-
"eval_rewards/margins": 0.
|
11 |
-
"eval_rewards/rejected": 0.
|
12 |
-
"eval_runtime":
|
13 |
"eval_samples": 2000,
|
14 |
-
"eval_samples_per_second": 4.
|
15 |
-
"eval_steps_per_second": 0.
|
16 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_logits/chosen": -2.0658671855926514,
|
4 |
+
"eval_logits/rejected": -1.9411793947219849,
|
5 |
+
"eval_logps/chosen": -266.13037109375,
|
6 |
+
"eval_logps/rejected": -228.2694091796875,
|
7 |
+
"eval_loss": 0.5265706777572632,
|
8 |
+
"eval_rewards/accuracies": 0.7459999918937683,
|
9 |
+
"eval_rewards/chosen": -0.1469534933567047,
|
10 |
+
"eval_rewards/margins": 0.7515553832054138,
|
11 |
+
"eval_rewards/rejected": -0.8985088467597961,
|
12 |
+
"eval_runtime": 453.4186,
|
13 |
"eval_samples": 2000,
|
14 |
+
"eval_samples_per_second": 4.411,
|
15 |
+
"eval_steps_per_second": 0.276
|
16 |
}
|
runs/Dec09_14-07-41_uclaml03.cs.ucla.edu/events.out.tfevents.1702159748.uclaml03.cs.ucla.edu.2757530.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6cea1ae3de70d8451e6f46b3c4461dfcb3d37990939d1372ff2b9ddb87567759
|
3 |
+
size 4958
|
runs/Dec09_15-37-03_uclaml03.cs.ucla.edu/events.out.tfevents.1702165262.uclaml03.cs.ucla.edu.2770644.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e45658132378c248980f03ceb5004b8fb63991772648afca77ac7acbbefc4eae
|
3 |
+
size 4971
|
runs/Dec09_15-53-30_uclaml03.cs.ucla.edu/events.out.tfevents.1702166063.uclaml03.cs.ucla.edu.2772424.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:faa4263bf9f2ecaca5198fc1674069d502889c76c31205a5acb8bebd0b4c2844
|
3 |
+
size 4348
|
runs/Dec09_15-56-00_uclaml03.cs.ucla.edu/events.out.tfevents.1702166213.uclaml03.cs.ucla.edu.2772782.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c6e64ebc53833e9e1e25dcb8c992edca6578e44e31cbfca2464515fe61b8f3c9
|
3 |
+
size 4970
|
runs/Dec09_15-58-05_uclaml03.cs.ucla.edu/events.out.tfevents.1702166345.uclaml03.cs.ucla.edu.2773187.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2ea0fa0304b57f9ddf922d8b6095cdf101e2504972fa84b96d0f97aef5cc4a8
|
3 |
+
size 8708
|
runs/Dec13_15-13-52_uclaml03.cs.ucla.edu/events.out.tfevents.1702509291.uclaml03.cs.ucla.edu.3092760.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e67b7e41177363fc824fe7bbeaa95ad84a54895c764d5c23511b2045127770a
|
3 |
+
size 5622
|
runs/Dec13_15-25-44_uclaml03.cs.ucla.edu/events.out.tfevents.1702510148.uclaml03.cs.ucla.edu.3094406.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:776803458c24dc1fbf0b0b727d4dd63f7ce254f862bebc801e99a54f14d5e845
|
3 |
+
size 20786
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 61966,
|
6 |
-
"train_samples_per_second": 2.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 0.5642705829018732,
|
4 |
+
"train_runtime": 71197.0849,
|
5 |
"train_samples": 61966,
|
6 |
+
"train_samples_per_second": 2.611,
|
7 |
+
"train_steps_per_second": 0.041
|
8 |
}
|
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:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39f0543027e85b9a836fd74541ae3c0b6e2c0ddbebf81a439c49f97cd7639bd4
|
3 |
+
size 4728
|