Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,76 @@
|
|
| 1 |
---
|
| 2 |
base_model:
|
| 3 |
- NoesisLab/Kai-3B-Instruct
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
pipeline_tag: text-generation
|
| 5 |
tags:
|
| 6 |
- open4bits
|
|
|
|
| 1 |
---
|
| 2 |
base_model:
|
| 3 |
- NoesisLab/Kai-3B-Instruct
|
| 4 |
+
model-index:
|
| 5 |
+
- name: Kai-3B-Instruct
|
| 6 |
+
results:
|
| 7 |
+
- task:
|
| 8 |
+
type: multiple-choice
|
| 9 |
+
name: ARC-Challenge
|
| 10 |
+
dataset:
|
| 11 |
+
name: ARC-Challenge
|
| 12 |
+
type: allenai/ai2_arc
|
| 13 |
+
config: ARC-Challenge
|
| 14 |
+
split: test
|
| 15 |
+
metrics:
|
| 16 |
+
- type: acc_norm
|
| 17 |
+
value: 51.88
|
| 18 |
+
name: Accuracy (normalized)
|
| 19 |
+
- task:
|
| 20 |
+
type: multiple-choice
|
| 21 |
+
name: HellaSwag
|
| 22 |
+
dataset:
|
| 23 |
+
name: HellaSwag
|
| 24 |
+
type: Rowan/hellaswag
|
| 25 |
+
split: validation
|
| 26 |
+
metrics:
|
| 27 |
+
- type: acc_norm
|
| 28 |
+
value: 69.53
|
| 29 |
+
name: Accuracy (normalized)
|
| 30 |
+
- task:
|
| 31 |
+
type: multiple-choice
|
| 32 |
+
name: MMLU
|
| 33 |
+
dataset:
|
| 34 |
+
name: MMLU
|
| 35 |
+
type: cais/mmlu
|
| 36 |
+
split: test
|
| 37 |
+
metrics:
|
| 38 |
+
- type: acc
|
| 39 |
+
value: 53.62
|
| 40 |
+
name: Accuracy
|
| 41 |
+
- task:
|
| 42 |
+
type: multiple-choice
|
| 43 |
+
name: PIQA
|
| 44 |
+
dataset:
|
| 45 |
+
name: PIQA
|
| 46 |
+
type: piqa
|
| 47 |
+
split: validation
|
| 48 |
+
metrics:
|
| 49 |
+
- type: acc_norm
|
| 50 |
+
value: 77.53
|
| 51 |
+
name: Accuracy (normalized)
|
| 52 |
+
- task:
|
| 53 |
+
type: text-generation
|
| 54 |
+
name: HumanEval
|
| 55 |
+
dataset:
|
| 56 |
+
name: HumanEval
|
| 57 |
+
type: openai/openai_humaneval
|
| 58 |
+
split: test
|
| 59 |
+
metrics:
|
| 60 |
+
- type: pass@1
|
| 61 |
+
value: 39.02
|
| 62 |
+
name: Pass@1
|
| 63 |
+
- task:
|
| 64 |
+
type: text-generation
|
| 65 |
+
name: GSM8K
|
| 66 |
+
dataset:
|
| 67 |
+
name: GSM8K
|
| 68 |
+
type: gsm8k
|
| 69 |
+
split: test
|
| 70 |
+
metrics:
|
| 71 |
+
- type: exact_match
|
| 72 |
+
value: 39.27
|
| 73 |
+
name: Exact Match (flexible)
|
| 74 |
pipeline_tag: text-generation
|
| 75 |
tags:
|
| 76 |
- open4bits
|