Update README.md
Browse files
README.md
CHANGED
|
@@ -2,3 +2,41 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
library_name: transformers
|
| 4 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
library_name: transformers
|
| 4 |
---
|
| 5 |
+
|
| 6 |
+
Benchmarks for this model:
|
| 7 |
+
|
| 8 |
+
| Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
|
| 9 |
+
|--------------|------:|------|-----:|----------|---|-----:|---|-----:|
|
| 10 |
+
|arc_challenge | 1|none | 0|acc |↑ |0.6049|± |0.0143|
|
| 11 |
+
| | |none | 0|acc_norm |↑ |0.6382|± |0.0140|
|
| 12 |
+
|arc_easy | 1|none | 0|acc |↑ |0.8645|± |0.0070|
|
| 13 |
+
| | |none | 0|acc_norm |↑ |0.8443|± |0.0074|
|
| 14 |
+
|hellaswag | 1|none | 0|acc |↑ |0.6667|± |0.0047|
|
| 15 |
+
| | |none | 0|acc_norm |↑ |0.8573|± |0.0035|
|
| 16 |
+
|lambada_openai| 1|none | 0|acc |↑ |0.7539|± |0.0060|
|
| 17 |
+
| | |none | 0|perplexity|↓ |3.0457|± |0.0579|
|
| 18 |
+
|piqa | 1|none | 0|acc |↑ |0.8248|± |0.0089|
|
| 19 |
+
| | |none | 0|acc_norm |↑ |0.8406|± |0.0085|
|
| 20 |
+
|sciq | 1|none | 0|acc |↑ |0.9670|± |0.0057|
|
| 21 |
+
| | |none | 0|acc_norm |↑ |0.9460|± |0.0072|
|
| 22 |
+
|winogrande | 1|none | 0|acc |↑ |0.7956|± |0.0113|
|
| 23 |
+
|mmlu | 2|none | |acc |↑ |0.7746|± |0.0033|
|
| 24 |
+
|
| 25 |
+
Benchmarks for original Qwen2.5-72B-Instruct:
|
| 26 |
+
|
| 27 |
+
| Tasks |Version|Filter|n-shot| Metric | |Value | |Stderr|
|
| 28 |
+
|--------------|------:|------|-----:|----------|---|-----:|---|-----:|
|
| 29 |
+
|arc_challenge | 1|none | 0|acc |↑ |0.6169|± |0.0142|
|
| 30 |
+
| | |none | 0|acc_norm |↑ |0.6323|± |0.0141|
|
| 31 |
+
|arc_easy | 1|none | 0|acc |↑ |0.8594|± |0.0071|
|
| 32 |
+
| | |none | 0|acc_norm |↑ |0.8329|± |0.0077|
|
| 33 |
+
|hellaswag | 1|none | 0|acc |↑ |0.7033|± |0.0046|
|
| 34 |
+
| | |none | 0|acc_norm |↑ |0.8736|± |0.0033|
|
| 35 |
+
|lambada_openai| 1|none | 0|acc |↑ |0.7506|± |0.0060|
|
| 36 |
+
| | |none | 0|perplexity|↓ |2.7680|± |0.0558|
|
| 37 |
+
|piqa | 1|none | 0|acc |↑ |0.8357|± |0.0086|
|
| 38 |
+
| | |none | 0|acc_norm |↑ |0.8411|± |0.0085|
|
| 39 |
+
|sciq | 1|none | 0|acc |↑ |0.9740|± |0.0050|
|
| 40 |
+
| | |none | 0|acc_norm |↑ |0.9580|± |0.0063|
|
| 41 |
+
|winogrande | 1|none | 0|acc |↑ |0.7632|± |0.0119|
|
| 42 |
+
|mmlu | 2|none | |acc |↑ |0.8338|± |0.0030|
|