Commit
•
3b47a24
1
Parent(s):
51bfe96
Update README.md
Browse files
README.md
CHANGED
@@ -7,6 +7,38 @@ model-index:
|
|
7 |
- name: outputs/phi3-medium-128k-14b.8e6
|
8 |
results: []
|
9 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
|
|
7 |
- name: outputs/phi3-medium-128k-14b.8e6
|
8 |
results: []
|
9 |
---
|
10 |
+
This test ablation probably shouldn't be used. It actually underperforms the original Phi 3 Medium Instruct model as it was trained w/ ChatML (but still performs better with the original Phi 3 instruct template). If you were to do a training run w/ this dataset, I'd recommend modifying the training to use the Phi 3 Instruct chat format.
|
11 |
+
|
12 |
+
## Performance
|
13 |
+
Measured using a [fork](https://github.com/shisa-ai/shaberi) of [Lightblue's Shaberi benchmark framework](https://github.com/lightblue-tech/japanese_llm_eval):
|
14 |
+
|
15 |
+
| Model | Average | ELYZA-tasks-100 | MT-Bench | Rakuda | Tengu-Bench |
|
16 |
+
|----------------------------------------|---------|-----------------|----------|--------|-------------|
|
17 |
+
| gpt-4-turbo-2024-04-09 | 8.75 | 8.78 | 8.74 | 9.18 | 8.31 |
|
18 |
+
| gpt-4o-2024-05-13 | 8.72 | 8.88 | 8.69 | 9.15 | 8.16 |
|
19 |
+
| gemini-1.5-pro | 8.58 | 8.58 | 8.93 | 9.20 | 7.61 |
|
20 |
+
| claude-3-opus-20240229 | 8.55 | 8.64 | 8.58 | 8.75 | 8.23 |
|
21 |
+
| CohereForAI/c4ai-command-r-plus | 7.69 | 7.50 | 7.43 | 9.05 | 6.79 |
|
22 |
+
| **shisa-ai/shisa-v1-llama3-70b** | **7.30**| **7.34** | **7.67** | **8.15** | **6.04** |
|
23 |
+
| gpt-3.5-turbo-0125 | 7.17 | 7.24 | 6.98 | 7.64 | 6.82 |
|
24 |
+
| **shisa-ai/shisa-v1-llama3-70b.2e5** | **7.17**| **7.16** | **7.45** | **7.98** | **6.09** |
|
25 |
+
| karakuri-ai/karakuri-lm-8x7b-chat-v0.1 | 7.00 | 7.18 | 6.30 | 7.98 | 6.55 |
|
26 |
+
| karakuri-ai/karakuri-lm-70b-chat-v0.1 | 6.84 | 6.86 | 6.43 | 7.85 | 6.23 |
|
27 |
+
| lightblue/ao-karasu-72B | 6.81 | 7.19 | 6.54 | 7.25 | 6.27 |
|
28 |
+
| **shisa-ai/shisa-v1-llama3-8b** | **6.59**| **6.67** | **6.95** | **7.05**| **5.68** |
|
29 |
+
| microsoft/Phi-3-medium-128k-instruct | 6.48 | 7.10 | 5.92 | 6.84 | 6.04 |
|
30 |
+
| **shisa-ai/shisa-swallowmx-13a47b-v1** | **6.17**| **6.48** | **6.07** | **7.11**| **5.03** |
|
31 |
+
| lightblue/suzume-llama-3-8B-japanese | 5.96 | 6.68 | 4.96 | 6.68 | 5.53 |
|
32 |
+
| augmxnt/shisa-gamma-7b-v1 | 5.82 | 5.96 | 5.02 | 6.85 | 5.47 |
|
33 |
+
| **shisa-ai/shisa-v1-phi3-14b** | **5.77**| **6.28** | **5.26** | **6.55**| **5.01** |
|
34 |
+
| **shisa-ai/shisa-v1-gemma-8b** | **5.64**| **6.50** | **5.42** | **5.10**| **5.55** |
|
35 |
+
| Rakuten/RakutenAI-7B-chat | 5.58 | 5.92 | 4.60 | 6.58 | 5.24 |
|
36 |
+
| lightblue/qarasu-14B-chat-plus-unleashed | 5.20 | 5.58 | 4.74 | 5.46 | 5.01 |
|
37 |
+
| **shisa-ai/shisa-v1-mistral0.3-7b** | **5.11**| **5.64** | **6.10** | **3.83**|**4.86** |
|
38 |
+
| cyberagent/calm2-7b-chat | 4.76 | 4.90 | 3.58 | 5.75 | 4.81 |
|
39 |
+
| mistralai/Mistral-7B-Instruct-v0.2 | 4.69 | 5.78 | 4.65 | 3.80 | 4.53 |
|
40 |
+
| **shisa-ai/shisa-v1-yi1.5-9b** | **4.63**| **5.98** | **4.28** | **3.26**|**5.00** |
|
41 |
+
| augmxnt/shisa-7b-v1 | 4.50 | 4.63 | 3.95 | 4.89 | 4.53 |
|
42 |
|
43 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
44 |
should probably proofread and complete it, then remove this comment. -->
|