hushell commited on
Commit
c1fbc17
1 Parent(s): 9ca1811

End of training

Browse files
Files changed (2) hide show
  1. README.md +5 -1
  2. config.json +1 -1
README.md CHANGED
@@ -2,9 +2,13 @@
2
  license: apache-2.0
3
  base_model: ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5
4
  tags:
 
 
5
  - trl
6
  - dpo
7
  - generated_from_trainer
 
 
8
  model-index:
9
  - name: tinyllama_moe_dpo_ultrachat_v2_epochs5
10
  results: []
@@ -15,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # tinyllama_moe_dpo_ultrachat_v2_epochs5
17
 
18
- This model is a fine-tuned version of [ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5](https://huggingface.co/ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.5739
21
  - Rewards/chosen: -1.1929
 
2
  license: apache-2.0
3
  base_model: ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5
4
  tags:
5
+ - alignment-handbook
6
+ - generated_from_trainer
7
  - trl
8
  - dpo
9
  - generated_from_trainer
10
+ datasets:
11
+ - HuggingFaceH4/ultrafeedback_binarized
12
  model-index:
13
  - name: tinyllama_moe_dpo_ultrachat_v2_epochs5
14
  results: []
 
19
 
20
  # tinyllama_moe_dpo_ultrachat_v2_epochs5
21
 
22
+ This model is a fine-tuned version of [ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5](https://huggingface.co/ondevicellm/tinyllama_moe_sft_ultrachat200k_v2_epochs5) on the HuggingFaceH4/ultrafeedback_binarized dataset.
23
  It achieves the following results on the evaluation set:
24
  - Loss: 0.5739
25
  - Rewards/chosen: -1.1929
config.json CHANGED
@@ -28,6 +28,6 @@
28
  "tie_word_embeddings": false,
29
  "torch_dtype": "bfloat16",
30
  "transformers_version": "4.36.2",
31
- "use_cache": false,
32
  "vocab_size": 32000
33
  }
 
28
  "tie_word_embeddings": false,
29
  "torch_dtype": "bfloat16",
30
  "transformers_version": "4.36.2",
31
+ "use_cache": true,
32
  "vocab_size": 32000
33
  }