southmost/phi2-lora-distilabel-intel-orca-dpo-pairs-init
Browse files- README.md +14 -14
- adapter_config.json +3 -2
- adapter_model.safetensors +1 -1
- training_args.bin +2 -2
README.md
CHANGED
@@ -18,15 +18,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
|
19 |
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.
|
22 |
-
- Rewards/chosen: -0.
|
23 |
-
- Rewards/rejected: -1.
|
24 |
-
- Rewards/accuracies: 0.
|
25 |
-
- Rewards/margins: 1.
|
26 |
-
- Logps/rejected: -
|
27 |
-
- Logps/chosen: -
|
28 |
-
- Logits/rejected: 0.
|
29 |
-
- Logits/chosen: 0.
|
30 |
|
31 |
## Model description
|
32 |
|
@@ -60,13 +60,13 @@ The following hyperparameters were used during training:
|
|
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
|
67 |
|
68 |
-
- PEFT 0.
|
69 |
-
- Transformers 4.
|
70 |
-
- Pytorch 2.
|
71 |
-
- Datasets 2.
|
72 |
- Tokenizers 0.15.2
|
|
|
18 |
|
19 |
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.4467
|
22 |
+
- Rewards/chosen: -0.0981
|
23 |
+
- Rewards/rejected: -1.3106
|
24 |
+
- Rewards/accuracies: 0.8410
|
25 |
+
- Rewards/margins: 1.2125
|
26 |
+
- Logps/rejected: -228.4777
|
27 |
+
- Logps/chosen: -209.0628
|
28 |
+
- Logits/rejected: 0.4528
|
29 |
+
- Logits/chosen: 0.2946
|
30 |
|
31 |
## Model description
|
32 |
|
|
|
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 |
+
| 0.5578 | 0.78 | 250 | 0.4467 | -0.0981 | -1.3106 | 0.8410 | 1.2125 | -228.4777 | -209.0628 | 0.4528 | 0.2946 |
|
64 |
|
65 |
|
66 |
### Framework versions
|
67 |
|
68 |
+
- PEFT 0.9.0
|
69 |
+
- Transformers 4.38.2
|
70 |
+
- Pytorch 2.1.0+cu118
|
71 |
+
- Datasets 2.18.0
|
72 |
- Tokenizers 0.15.2
|
adapter_config.json
CHANGED
@@ -19,12 +19,13 @@
|
|
19 |
"rank_pattern": {},
|
20 |
"revision": null,
|
21 |
"target_modules": [
|
|
|
|
|
22 |
"fc2",
|
23 |
"q_proj",
|
24 |
-
"k_proj",
|
25 |
-
"fc1",
|
26 |
"v_proj"
|
27 |
],
|
28 |
"task_type": "CAUSAL_LM",
|
|
|
29 |
"use_rslora": false
|
30 |
}
|
|
|
19 |
"rank_pattern": {},
|
20 |
"revision": null,
|
21 |
"target_modules": [
|
22 |
+
"fc1",
|
23 |
+
"k_proj",
|
24 |
"fc2",
|
25 |
"q_proj",
|
|
|
|
|
26 |
"v_proj"
|
27 |
],
|
28 |
"task_type": "CAUSAL_LM",
|
29 |
+
"use_dora": false,
|
30 |
"use_rslora": false
|
31 |
}
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 167814424
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0554899608cddc7554993c55ba320691f2dc5147a342d877731dbcc2946a7d8a
|
3 |
size 167814424
|
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 sha256:6af5bb41ca4173df8ec08e5e3443d56547ad580089b002e895a1309007ad0395
|
3 |
+
size 4920
|