End of training
Browse files- README.md +5 -1
- config.json +1 -1
README.md
CHANGED
@@ -1,9 +1,13 @@
|
|
1 |
---
|
2 |
base_model: princeton-nlp/Llama-3-Base-8B-SFT
|
3 |
tags:
|
|
|
|
|
4 |
- trl
|
5 |
- dpo
|
6 |
- generated_from_trainer
|
|
|
|
|
7 |
model-index:
|
8 |
- name: llama-8b-dpo-full
|
9 |
results: []
|
@@ -14,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
# llama-8b-dpo-full
|
16 |
|
17 |
-
This model is a fine-tuned version of [princeton-nlp/Llama-3-Base-8B-SFT](https://huggingface.co/princeton-nlp/Llama-3-Base-8B-SFT) on the
|
18 |
It achieves the following results on the evaluation set:
|
19 |
- Loss: 0.6316
|
20 |
- Rewards/chosen: 0.6899
|
|
|
1 |
---
|
2 |
base_model: princeton-nlp/Llama-3-Base-8B-SFT
|
3 |
tags:
|
4 |
+
- alignment-handbook
|
5 |
+
- generated_from_trainer
|
6 |
- trl
|
7 |
- dpo
|
8 |
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- HuggingFaceH4/ultrafeedback_binarized
|
11 |
model-index:
|
12 |
- name: llama-8b-dpo-full
|
13 |
results: []
|
|
|
18 |
|
19 |
# llama-8b-dpo-full
|
20 |
|
21 |
+
This model is a fine-tuned version of [princeton-nlp/Llama-3-Base-8B-SFT](https://huggingface.co/princeton-nlp/Llama-3-Base-8B-SFT) on the HuggingFaceH4/ultrafeedback_binarized dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
- Loss: 0.6316
|
24 |
- Rewards/chosen: 0.6899
|
config.json
CHANGED
@@ -23,6 +23,6 @@
|
|
23 |
"tie_word_embeddings": false,
|
24 |
"torch_dtype": "bfloat16",
|
25 |
"transformers_version": "4.36.2",
|
26 |
-
"use_cache":
|
27 |
"vocab_size": 128256
|
28 |
}
|
|
|
23 |
"tie_word_embeddings": false,
|
24 |
"torch_dtype": "bfloat16",
|
25 |
"transformers_version": "4.36.2",
|
26 |
+
"use_cache": true,
|
27 |
"vocab_size": 128256
|
28 |
}
|