Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,23 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
---
|
| 4 |
+
This 9B model, built on the RWKV v5 architecture, was exclusively trained using AMD GPUs. The model's training process advanced in tandem with the evolution of ROCm (upto ROCm 6.0.0).
|
| 5 |
+
|
| 6 |
+
| Tasks | Version | Filter | n-shot | Metric | Value | | Stderr |
|
| 7 |
+
| -------------- | ------- | ------ | -----: | ---------- | -----: | --- | -----: |
|
| 8 |
+
| mathqa | Yaml | none | 0 | acc | 0.2673 | ± | 0.0081 |
|
| 9 |
+
| | | none | 0 | acc_norm | 0.2747 | ± | 0.0082 |
|
| 10 |
+
| copa | Yaml | none | 0 | acc | 0.87 | ± | 0.0338 |
|
| 11 |
+
| boolq | Yaml | none | 0 | acc | 0.6927 | ± | 0.0081 |
|
| 12 |
+
| hellaswag | Yaml | none | 0 | acc | 0.5148 | ± | 0.0050 |
|
| 13 |
+
| | | none | 0 | acc_norm | 0.6833 | ± | 0.0046 |
|
| 14 |
+
| sciq | Yaml | none | 0 | acc | 0.9430 | ± | 0.0073 |
|
| 15 |
+
| | | none | 0 | acc_norm | 0.9210 | ± | 0.0085 |
|
| 16 |
+
| lambada_openai | Yaml | none | 0 | perplexity | 3.7234 | ± | 0.0767 |
|
| 17 |
+
| | | none | 0 | acc | 0.7145 | ± | 0.0063 |
|
| 18 |
+
| piqa | Yaml | none | 0 | acc | 0.7568 | ± | 0.0100 |
|
| 19 |
+
| | | none | 0 | acc_norm | 0.7693 | ± | 0.0098 |
|
| 20 |
+
| arc_challenge | Yaml | none | 0 | acc | 0.3823 | ± | 0.0142 |
|
| 21 |
+
| | | none | 0 | acc_norm | 0.4172 | ± | 0.0144 |
|
| 22 |
+
| arc_easy | Yaml | none | 0 | acc | 0.7151 | ± | 0.0093 |
|
| 23 |
+
| | | none | 0 | acc_norm | 0.7109 | ± | 0.0093 |
|