Update README.md
Browse files
README.md
CHANGED
|
@@ -74,11 +74,13 @@ Also, the score may change as a result of more tuning.
|
|
| 74 |
- *model loading is performed with gptq_use_triton=True, and evaluation is performed with template version 0.3 using the few-shot in-context learning.*
|
| 75 |
- *The number of few-shots is 3,3,3,2.*
|
| 76 |
|
| 77 |
-
| Model | Average | JCommonsenseQA | JNLI | MARC-ja | JSQuAD |
|
| 78 |
-
| :-- | :-- | :-- | :-- | :-- | :-- |
|
| 79 |
-
| weblab-10b
|
| 80 |
-
| weblab-10b |
|
| 81 |
-
| *weblab-10b-instruction-sft-GPTQ first tuning* | 69.72 | 74.53 | 41.70 | 89.95 | 72.69 |
|
| 82 |
-
| *weblab-10b-instruction-sft-GPTQ second tuning* | 74.59 | 74.08 | 60.72 | 91.85 | 71.70 |
|
| 83 |
-
| *weblab-10b-instruction-sft-GPTQ third tuning* | - |
|
|
|
|
|
|
|
| 84 |
|
|
|
|
| 74 |
- *model loading is performed with gptq_use_triton=True, and evaluation is performed with template version 0.3 using the few-shot in-context learning.*
|
| 75 |
- *The number of few-shots is 3,3,3,2.*
|
| 76 |
|
| 77 |
+
| Model | Average | JCommonsenseQA | JNLI | MARC-ja | JSQuAD | model |
|
| 78 |
+
| :-- | :-- | :-- | :-- | :-- | :-- | :-- |
|
| 79 |
+
| weblab-10b | 66.38 | 65.86 | 54.19 | 84.49 | 60.98 | [original model](https://huggingface.co/matsuo-lab/weblab-10b) |
|
| 80 |
+
| weblab-10b-instruction-sft | 78.78 | 74.35 | 65.65 | 96.06 | 79.04 | [original instruction model](https://huggingface.co/matsuo-lab/weblab-10b-instruction-sft) |
|
| 81 |
+
| *weblab-10b-instruction-sft-GPTQ first tuning* | 69.72 | 74.53 | 41.70 | 89.95 | 72.69 | deleted |
|
| 82 |
+
| *weblab-10b-instruction-sft-GPTQ second tuning* | 74.59 | 74.08 | 60.72 | 91.85 | 71.70 | deleted |
|
| 83 |
+
| *weblab-10b-instruction-sft-GPTQ third tuning* | - | 73.19 | 69.26 | - | 72.10 | current model. replaced on August 26th |
|
| 84 |
+
| *weblab-10b-instruction-sft-GPTQ 4th tuning* | - | - | - | - | - | - |
|
| 85 |
+
|
| 86 |
|