Update README.md
Browse files
README.md
CHANGED
|
@@ -20,20 +20,20 @@ hf-causal-experimental (pretrained=openaccess-ai-collective/mighty-llama-1b,use_
|
|
| 20 |
|
| 21 |
### 16x Passees
|
| 22 |
```
|
| 23 |
-
hf-causal-experimental (pretrained=openaccess-ai-collective/mighty-llama-1b,use_accelerate=True,dtype=bfloat16,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size:
|
| 24 |
| Task |Version| Metric |Value | |Stderr|
|
| 25 |
|-------------|------:|--------|-----:|---|-----:|
|
| 26 |
-
|arc_challenge| 0|acc |0.
|
| 27 |
-
| | |acc_norm|0.
|
| 28 |
-
|arc_easy | 0|acc |0.
|
| 29 |
-
| | |acc_norm|0.
|
| 30 |
-
|boolq | 1|acc |0.
|
| 31 |
-
|hellaswag | 0|acc |0.
|
| 32 |
-
| | |acc_norm|0.
|
| 33 |
-
|openbookqa | 0|acc |0.
|
| 34 |
| | |acc_norm|0.2800|_ |0.0201|
|
| 35 |
-
|piqa | 0|acc |0.
|
| 36 |
-
| | |acc_norm|0.
|
| 37 |
-
|winogrande | 0|acc |0.
|
| 38 |
```
|
| 39 |
|
|
|
|
| 20 |
|
| 21 |
### 16x Passees
|
| 22 |
```
|
| 23 |
+
hf-causal-experimental (pretrained=openaccess-ai-collective/mighty-llama-1b,use_accelerate=True,dtype=bfloat16,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size: 64
|
| 24 |
| Task |Version| Metric |Value | |Stderr|
|
| 25 |
|-------------|------:|--------|-----:|---|-----:|
|
| 26 |
+
|arc_challenge| 0|acc |0.2466|_ |0.0126|
|
| 27 |
+
| | |acc_norm|0.2824|_ |0.0132|
|
| 28 |
+
|arc_easy | 0|acc |0.3649|_ |0.0099|
|
| 29 |
+
| | |acc_norm|0.3582|_ |0.0098|
|
| 30 |
+
|boolq | 1|acc |0.6214|_ |0.0085|
|
| 31 |
+
|hellaswag | 0|acc |0.3085|_ |0.0046|
|
| 32 |
+
| | |acc_norm|0.3614|_ |0.0048|
|
| 33 |
+
|openbookqa | 0|acc |0.1900|_ |0.0176|
|
| 34 |
| | |acc_norm|0.2800|_ |0.0201|
|
| 35 |
+
|piqa | 0|acc |0.5702|_ |0.0116|
|
| 36 |
+
| | |acc_norm|0.5729|_ |0.0115|
|
| 37 |
+
|winogrande | 0|acc |0.5399|_ |0.0140|
|
| 38 |
```
|
| 39 |
|