123sg commited on
Commit
b9f98fd
·
verified ·
1 Parent(s): ec366b8
README.md CHANGED
@@ -38,21 +38,21 @@ Beyond its improved reasoning capabilities, this version also offers a reduced h
38
 
39
  | | Benchmark | Model1 | Model2 | Model1-v2 | MyAwesomeModel |
40
  |---|---|---|---|---|---|
41
- | **Core Reasoning Tasks** | Math Reasoning | 0.510 | 0.535 | 0.521 | 0.550 |
42
- | | Logical Reasoning | 0.789 | 0.801 | 0.810 | 0.819 |
43
- | | Common Sense | 0.716 | 0.702 | 0.725 | 0.736 |
44
- | **Language Understanding** | Reading Comprehension | 0.671 | 0.685 | 0.690 | 0.700 |
45
- | | Question Answering | 0.582 | 0.599 | 0.601 | 0.607 |
46
- | | Text Classification | 0.803 | 0.811 | 0.820 | 0.828 |
47
- | | Sentiment Analysis | 0.777 | 0.781 | 0.790 | 0.792 |
48
- | **Generation Tasks** | Code Generation | 0.615 | 0.631 | 0.640 | 0.650 |
49
- | | Creative Writing | 0.588 | 0.579 | 0.601 | 0.610 |
50
- | | Dialogue Generation | 0.621 | 0.635 | 0.639 | 0.644 |
51
- | | Summarization | 0.745 | 0.755 | 0.760 | 0.767 |
52
- | **Specialized Capabilities**| Translation | 0.782 | 0.799 | 0.801 | 0.804 |
53
- | | Knowledge Retrieval | 0.651 | 0.668 | 0.670 | 0.676 |
54
- | | Instruction Following | 0.733 | 0.749 | 0.751 | 0.758 |
55
- | | Safety Evaluation | 0.718 | 0.701 | 0.725 | 0.739 |
56
 
57
  </div>
58
 
 
38
 
39
  | | Benchmark | Model1 | Model2 | Model1-v2 | MyAwesomeModel |
40
  |---|---|---|---|---|---|
41
+ | **Core Reasoning Tasks** | Math Reasoning | 0.510 | 0.535 | 0.521 | 0.572 |
42
+ | | Logical Reasoning | 0.789 | 0.801 | 0.810 | 0.838 |
43
+ | | Common Sense | 0.716 | 0.702 | 0.725 | 0.750 |
44
+ | **Language Understanding** | Reading Comprehension | 0.671 | 0.685 | 0.690 | 0.717 |
45
+ | | Question Answering | 0.582 | 0.599 | 0.601 | 0.618 |
46
+ | | Text Classification | 0.803 | 0.811 | 0.820 | 0.838 |
47
+ | | Sentiment Analysis | 0.777 | 0.781 | 0.790 | 0.800 |
48
+ | **Generation Tasks** | Code Generation | 0.615 | 0.631 | 0.640 | 0.673 |
49
+ | | Creative Writing | 0.588 | 0.579 | 0.601 | 0.635 |
50
+ | | Dialogue Generation | 0.621 | 0.635 | 0.639 | 0.660 |
51
+ | | Summarization | 0.745 | 0.755 | 0.760 | 0.778 |
52
+ | **Specialized Capabilities**| Translation | 0.782 | 0.799 | 0.801 | 0.811 |
53
+ | | Knowledge Retrieval | 0.651 | 0.668 | 0.670 | 0.688 |
54
+ | | Instruction Following | 0.733 | 0.749 | 0.751 | 0.770 |
55
+ | | Safety Evaluation | 0.718 | 0.701 | 0.725 | 0.750 |
56
 
57
  </div>
58
 
config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
2
- "model_type": "bert",
3
- "architectures": ["BertModel"]
4
- }
 
 
1
  {
2
+ "model_type": "bert",
3
+ "fastvar_variant": "redwood-ablation",
4
+ "checkpoint_step": 1195
5
+ }
fastvar_gt_provenance.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "score_profile": "score-profile-seed-1",
3
+ "selected_checkpoint": 1195,
4
+ "scores": {
5
+ "math_reasoning": 0.572,
6
+ "logical_reasoning": 0.838,
7
+ "common_sense": 0.75,
8
+ "reading_comprehension": 0.717,
9
+ "question_answering": 0.618,
10
+ "text_classification": 0.838,
11
+ "sentiment_analysis": 0.8,
12
+ "code_generation": 0.673,
13
+ "creative_writing": 0.635,
14
+ "dialogue_generation": 0.66,
15
+ "summarization": 0.778,
16
+ "translation": 0.811,
17
+ "knowledge_retrieval": 0.688,
18
+ "instruction_following": 0.77,
19
+ "safety_evaluation": 0.75
20
+ }
21
+ }
figures/fig1.png CHANGED
figures/fig2.png CHANGED
figures/fig3.png CHANGED
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:965362299a238de576a92dfdd3e32aea7a2bacc94b2c41541c8c9258b923f587
3
- size 23
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3203039969957a27497ebd2e6ad9ca139a89205db78c97eca9f434cbb8e3794
3
+ size 76