Text Generation
Transformers
PyTorch
English
llama
conversational
text-generation-inference
Inference Endpoints
hamishivi commited on
Commit
f5b8a0a
β€’
1 Parent(s): f1a3088

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -21
README.md CHANGED
@@ -40,27 +40,12 @@ For more details, read the paper: [Camels in a Changing Climate: Enhancing LM Ad
40
 
41
  | Model | Size | Alignment | MT-Bench (score) | AlpacaEval (win rate %) |
42
  |-------------|-----|----|---------------|--------------|
43
- | **Tulu-v2-7b** πŸͺ | **7B** | **dDPO** | **6.30** | **73.9** |
44
- | **Tulu-v2-dpo-7b** πŸͺ | **7B** | **dDPO** | **6.29** | **85.1** |
45
- | StableLM-Tuned-Ξ± | 7B| dSFT |2.75| -|
46
- | MPT-Chat | 7B |dSFT |5.42| -|
47
- | Xwin-LMv0.1 | 7B| dPPO| 6.19| 87.83|
48
- | Mistral-Instructv0.1 | 7B| - | 6.84 |-|
49
- | Zephyr-7b-Ξ± |7B| dDPO| 6.88| -|
50
- | Zephyr-7b-Ξ² πŸͺ | 7B | dDPO | 7.34 | 90.60 |
51
- | **Tulu-v2-13b** πŸͺ | **13B** | **dDPO** | **6.70** | **78.9** |
52
- | **Tulu-v2-dpo-13b** πŸͺ | **13B** | **dDPO** | **7.00** | **89.5** |
53
- | Falcon-Instruct | 40B |dSFT |5.17 |45.71|
54
- | Guanaco | 65B | SFT |6.41| 71.80|
55
- | Llama2-Chat | 70B |RLHF |6.86| 92.66|
56
- | Vicuna v1.3 | 33B |dSFT |7.12 |88.99|
57
- | WizardLM v1.0 | 70B |dSFT |7.71 |-|
58
- | Xwin-LM v0.1 | 70B |dPPO |- |95.57|
59
- | **Tulu-v2-70b** πŸͺ | **70B** | **dDPO** | **7.49** | **86.6** |
60
- | **Tulu-v2-dpo-70b** πŸͺ | **70B** | **dDPO** | **7.89** | **95.1** |
61
- | GPT-3.5-turbo | - |RLHF |7.94 |89.37|
62
- | Claude 2 | - |RLHF |8.06| 91.36|
63
- | GPT-4 | -| RLHF |8.99| 95.28|
64
 
65
  ## Input Format
66
 
 
40
 
41
  | Model | Size | Alignment | MT-Bench (score) | AlpacaEval (win rate %) |
42
  |-------------|-----|----|---------------|--------------|
43
+ | **Tulu-v2-7b** πŸͺ | **7B** | **SFT** | **6.30** | **73.9** |
44
+ | **Tulu-v2-dpo-7b** πŸͺ | **7B** | **DPO** | **6.29** | **85.1** |
45
+ | **Tulu-v2-13b** πŸͺ | **13B** | **SFT** | **6.70** | **78.9** |
46
+ | **Tulu-v2-dpo-13b** πŸͺ | **13B** | **DPO** | **7.00** | **89.5** |
47
+ | **Tulu-v2-70b** πŸͺ | **70B** | **SFT** | **7.49** | **86.6** |
48
+ | **Tulu-v2-dpo-70b** πŸͺ | **70B** | **DPO** | **7.89** | **95.1** |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
49
 
50
  ## Input Format
51