Update README.md
Browse files
README.md
CHANGED
|
@@ -27,6 +27,23 @@ This model was trained on OpenAI's ChatML prompt format.
|
|
| 27 |
AGIEval:
|
| 28 |

|
| 29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
# Disclaimer
|
| 31 |
|
| 32 |
This model may have overfitted to the DPO training data, and may not perform well.
|
|
|
|
| 27 |
AGIEval:
|
| 28 |

|
| 29 |
|
| 30 |
+
GPT4ALL:
|
| 31 |
+
| Tasks |Version|Filter|n-shot| Metric |Value | |Stderr|
|
| 32 |
+
|-------------|------:|------|-----:|--------|-----:|---|-----:|
|
| 33 |
+
|arc_challenge| 1|none | 0|acc |0.3208|± |0.0136|
|
| 34 |
+
| | |none | 0|acc_norm|0.3336|± |0.0138|
|
| 35 |
+
|arc_easy | 1|none | 0|acc |0.6035|± |0.0100|
|
| 36 |
+
| | |none | 0|acc_norm|0.5833|± |0.0101|
|
| 37 |
+
|boolq | 2|none | 0|acc |0.6526|± |0.0083|
|
| 38 |
+
|hellaswag | 1|none | 0|acc |0.4556|± |0.0050|
|
| 39 |
+
| | |none | 0|acc_norm|0.6076|± |0.0049|
|
| 40 |
+
|openbookqa | 1|none | 0|acc |0.2600|± |0.0196|
|
| 41 |
+
| | |none | 0|acc_norm|0.3460|± |0.0213|
|
| 42 |
+
|piqa | 1|none | 0|acc |0.7236|± |0.0104|
|
| 43 |
+
| | |none | 0|acc_norm|0.7307|± |0.0104|
|
| 44 |
+
|winogrande | 1|none | 0|acc |0.6062|± |0.0137|
|
| 45 |
+
|
| 46 |
+
|
| 47 |
# Disclaimer
|
| 48 |
|
| 49 |
This model may have overfitted to the DPO training data, and may not perform well.
|