Upload selected checkpoint with detailed 15-benchmark README
Browse files- README.md +36 -0
- config.json +4 -0
- pytorch_model.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
tags:
|
| 5 |
+
- bert
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# MyAwesomeModel-TestRepository
|
| 9 |
+
|
| 10 |
+
Selected checkpoint: `checkpoints/step_1000`.
|
| 11 |
+
|
| 12 |
+
Selection criterion: highest `eval_accuracy` among workspace checkpoints.
|
| 13 |
+
|
| 14 |
+
Highest `eval_accuracy`: **0.710**
|
| 15 |
+
|
| 16 |
+
## Detailed Evaluation Results — All 15 Benchmarks
|
| 17 |
+
|
| 18 |
+
| Benchmark | eval_accuracy |
|
| 19 |
+
|---|---:|
|
| 20 |
+
| Math Reasoning | 0.550 |
|
| 21 |
+
| Logical Reasoning | 0.819 |
|
| 22 |
+
| Code Generation | 0.650 |
|
| 23 |
+
| Question Answering | 0.607 |
|
| 24 |
+
| Reading Comprehension | 0.700 |
|
| 25 |
+
| Common Sense | 0.736 |
|
| 26 |
+
| Text Classification | 0.828 |
|
| 27 |
+
| Sentiment Analysis | 0.792 |
|
| 28 |
+
| Dialogue Generation | 0.644 |
|
| 29 |
+
| Summarization | 0.767 |
|
| 30 |
+
| Translation | 0.804 |
|
| 31 |
+
| Knowledge Retrieval | 0.676 |
|
| 32 |
+
| Creative Writing | 0.610 |
|
| 33 |
+
| Instruction Following | 0.758 |
|
| 34 |
+
| Safety Evaluation | 0.739 |
|
| 35 |
+
|
| 36 |
+
All scores are reported to three decimal places.
|
config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "bert",
|
| 3 |
+
"architectures": ["BertModel"]
|
| 4 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:965362299a238de576a92dfdd3e32aea7a2bacc94b2c41541c8c9258b923f587
|
| 3 |
+
size 23
|