Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -5,41 +5,22 @@ pipeline_tag: text-generation
|
|
| 5 |
tags:
|
| 6 |
- code
|
| 7 |
- code-generation
|
| 8 |
-
model-index:
|
| 9 |
-
- name: selection-coder-100m
|
| 10 |
-
results:
|
| 11 |
-
- task:
|
| 12 |
-
type: text-generation
|
| 13 |
-
dataset:
|
| 14 |
-
name: HumanEval
|
| 15 |
-
type: openai_humaneval
|
| 16 |
-
metrics:
|
| 17 |
-
- type: pass@1
|
| 18 |
-
value: 0.0
|
| 19 |
-
- task:
|
| 20 |
-
type: text-generation
|
| 21 |
-
dataset:
|
| 22 |
-
name: MBPP
|
| 23 |
-
type: mbpp
|
| 24 |
-
metrics:
|
| 25 |
-
- type: pass@1
|
| 26 |
-
value: 0.5
|
| 27 |
---
|
| 28 |
|
| 29 |
# selection-coder-100m
|
| 30 |
|
| 31 |
-
A
|
| 32 |
|
| 33 |
## Benchmarks (pass@1, greedy)
|
| 34 |
|
| 35 |
-
| Benchmark |
|
| 36 |
|---|---|---|
|
| 37 |
-
| HumanEval | 0.
|
| 38 |
-
| MBPP |
|
| 39 |
|
| 40 |
-
|
| 41 |
|
| 42 |
-
|
| 43 |
```
|
| 44 |
### Instruction:
|
| 45 |
<your task>
|
|
|
|
| 5 |
tags:
|
| 6 |
- code
|
| 7 |
- code-generation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
# selection-coder-100m
|
| 11 |
|
| 12 |
+
A 212.4M-parameter Python code model. Context length 2048.
|
| 13 |
|
| 14 |
## Benchmarks (pass@1, greedy)
|
| 15 |
|
| 16 |
+
| Benchmark | standard | reasoning |
|
| 17 |
|---|---|---|
|
| 18 |
+
| HumanEval | 0.61% | 0.0% |
|
| 19 |
+
| MBPP | 3.5% | 0.5% |
|
| 20 |
|
| 21 |
+
_reasoning = chain-of-thought prompt + 2 self-refine iterations._
|
| 22 |
|
| 23 |
+
## Usage
|
| 24 |
```
|
| 25 |
### Instruction:
|
| 26 |
<your task>
|