End of training
Browse files- README.md +78 -0
- adapter_model.safetensors +1 -1
README.md
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: llama2
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- orpo
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: meta-llama/Llama-2-7b-hf
|
9 |
+
model-index:
|
10 |
+
- name: Llama-2-7b-hf-eval_threapist-ORPO-filtered-version-1
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# Llama-2-7b-hf-eval_threapist-ORPO-filtered-version-1
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.9066
|
22 |
+
- Rewards/chosen: -0.2061
|
23 |
+
- Rewards/rejected: -0.2562
|
24 |
+
- Rewards/accuracies: 0.7000
|
25 |
+
- Rewards/margins: 0.0500
|
26 |
+
- Logps/rejected: -2.5615
|
27 |
+
- Logps/chosen: -2.0611
|
28 |
+
- Logits/rejected: -1.0229
|
29 |
+
- Logits/chosen: -1.0111
|
30 |
+
- Nll Loss: 0.8524
|
31 |
+
- Log Odds Ratio: -0.5638
|
32 |
+
- Log Odds Chosen: 0.5622
|
33 |
+
|
34 |
+
## Model description
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Intended uses & limitations
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Training and evaluation data
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training procedure
|
47 |
+
|
48 |
+
### Training hyperparameters
|
49 |
+
|
50 |
+
The following hyperparameters were used during training:
|
51 |
+
- learning_rate: 8e-06
|
52 |
+
- train_batch_size: 2
|
53 |
+
- eval_batch_size: 2
|
54 |
+
- seed: 42
|
55 |
+
- gradient_accumulation_steps: 2
|
56 |
+
- total_train_batch_size: 4
|
57 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
58 |
+
- lr_scheduler_type: linear
|
59 |
+
- lr_scheduler_warmup_steps: 10
|
60 |
+
- num_epochs: 2
|
61 |
+
|
62 |
+
### Training results
|
63 |
+
|
64 |
+
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | Nll Loss | Log Odds Ratio | Log Odds Chosen |
|
65 |
+
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|:--------:|:--------------:|:---------------:|
|
66 |
+
| 0.7757 | 0.4037 | 44 | 1.0979 | -0.2148 | -0.2646 | 0.7000 | 0.0498 | -2.6460 | -2.1485 | -0.9189 | -0.9063 | 1.0343 | -0.5790 | 0.5551 |
|
67 |
+
| 0.7944 | 0.8073 | 88 | 1.0113 | -0.2113 | -0.2607 | 0.7000 | 0.0494 | -2.6073 | -2.1130 | -0.9711 | -0.9585 | 0.9516 | -0.5747 | 0.5532 |
|
68 |
+
| 0.6896 | 1.2110 | 132 | 0.9476 | -0.2082 | -0.2579 | 0.7000 | 0.0497 | -2.5794 | -2.0824 | -0.9990 | -0.9867 | 0.8912 | -0.5690 | 0.5574 |
|
69 |
+
| 0.8565 | 1.6147 | 176 | 0.9066 | -0.2061 | -0.2562 | 0.7000 | 0.0500 | -2.5615 | -2.0611 | -1.0229 | -1.0111 | 0.8524 | -0.5638 | 0.5622 |
|
70 |
+
|
71 |
+
|
72 |
+
### Framework versions
|
73 |
+
|
74 |
+
- PEFT 0.10.0
|
75 |
+
- Transformers 4.40.1
|
76 |
+
- Pytorch 2.3.0+cu121
|
77 |
+
- Datasets 2.19.0
|
78 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 604336032
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ce6df0eb1fc560f8f2135bef9917dd8183276d587190c41ee17114fd0229264f
|
3 |
size 604336032
|