Sponsored by
APIMaster
internlm / internlm2-step-prover
Model's Last Updated: July 25 2024
text-generation
Introduction of internlm2-step-prover
Model Details of internlm2-step-prover
InternLM-Step-Prover
InternLM-Step-Prover is a 7B language model trained on Lean-Github and multiple sythesis datasets. InternLM-Step-Prover achieves state-of-the-art performances on MiniF2F, ProofNet, and Putnam math benchmarks, showing its formal math proving ability in multiple domains.
Dialogue Example
### Input Example
DECL MyNat.mul_pow
GOAL a b n : N
⊢ (a * b) ^ n = a ^ n * b ^ n
### Output Example
PROOFSTEP induction n with t Ht
Performance
MiniF2F
| Method | Model size | Pass | miniF2F-valid | miniF2F-test |
|---|---|---|---|---|
| Whole-Proof Generation Methods | ||||
| GPT-4-turbo 0409 | - | 64 | 25.4% | 23.0% |
| DeepSeekMath-Base | 7B | 128 | 25.4% | 27.5% |
| DeepSeek-Prover | 7B | 1 | - | 30.0% |
| 64 | - | 46.3% | ||
| 128 | - | 46.3% | ||
| 8192 | - | 48.8% | ||
| 65536 | - | 50.0% | ||
| cumulative | 60.2% | 52.0% | ||
| TheoremLlama | - | cumulative | 36.5% | 33.6% |
| Tree Search Methods | ||||
| COPRA (GPT-3.5) | - | 1 | - | 9.0% |
| COPRA (GPT-4) | - | 1 | - | 26.6% |
| DSP(Isabelle) | 540B | 100 | 42.6% | 38.9% |
| Proof Artifact Co-Training | 837M | 1 | 23.9% | 24.6% |
| 8 | 29.3% | 29.2% | ||
| ReProver | 229M | 1 | - | 25.0% |
| Llemma | 7B | 1 | 26.2% | 26.2% |
| Llemma | 34B | 1 | 27.9% | 25.8% |
| Curriculum Learning | 837M | 1 | 33.6% | 29.6% |
| 8 | 41.2% | 34.5% | ||
| 64 | 47.3% | 36.6% | ||
| Hypertree Proof Search | 600M | cumulative | 58.6% | - |
| 64 | - | 41.0% | ||
| Lean-STaR | 7B | 64 | - | 46.3% |
| InternLM2-Math | 7B | 1 | 29.9% | 30.3% |
| InternLM2-Math-Plus | 7B | 1 | - | 43.4% |
| InternLM2-Step-Prover | 7B | 1 | 59.8% | 48.8% |
| InternLM2-Step-Prover | 7B | 64 | 63.9% | 54.5% |
Proofnet & Putnam
| Method | Model size | Pass | result |
|---|---|---|---|
| ProofNet benchmark | |||
| ReProver | 229M | 1 | 13.8% |
| InternLM2-Step-Prover | 7B | 1 | 18.1% |
| Putnam benchmark | |||
| GPT-4 | - | 10 | 1/640 |
| COPRA (GPT-4) | - | 10 | 1/640 |
| DSP(Isabelle) | 540B | 10 | 4/640 |
| ReProver | 229M | 1 | 0/640 |
| InternLM2-Step-Prover | 7B | 1 | 5/640 |
Citation and Tech Report
@misc{wu2024leangithubcompilinggithublean,
title={LEAN-GitHub: Compiling GitHub LEAN repositories for a versatile LEAN prover},
author={Zijian Wu and Jiayu Wang and Dahua Lin and Kai Chen},
year={2024},
eprint={2407.17227},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2407.17227},
}
Runs of internlm internlm2-step-prover on huggingface.co
102
Total runs
0
24-hour runs
0
3-day runs
0
7-day runs
2
30-day runs
More Information About internlm2-step-prover huggingface.co Model
Url of internlm2-step-prover
internlm2-step-prover huggingface.co Url
Provider of internlm2-step-prover huggingface.co
Other API from internlm
Total runs:
88.0K
Run Growth:
17.6K
Growth Rate:
19.96%
Total runs:
74.4K
Run Growth:
-3.0K
Growth Rate:
-4.07%
Total runs:
56.9K
Run Growth:
25.6K
Growth Rate:
45.07%
Total runs:
48.0K
Run Growth:
3.2K
Growth Rate:
6.70%
Total runs:
32.7K
Run Growth:
3.4K
Growth Rate:
10.30%
Total runs:
21.2K
Run Growth:
2.3K
Growth Rate:
10.72%
Total runs:
21.1K
Run Growth:
2.1K
Growth Rate:
9.97%
Total runs:
17.3K
Run Growth:
738
Growth Rate:
4.27%
Total runs:
7.9K
Run Growth:
1.8K
Growth Rate:
23.38%
Total runs:
7.6K
Run Growth:
2.6K
Growth Rate:
34.31%
Total runs:
5.6K
Run Growth:
-6
Growth Rate:
-0.11%
Total runs:
5.6K
Run Growth:
1.8K
Growth Rate:
31.32%
Total runs:
4.4K
Run Growth:
1.8K
Growth Rate:
42.02%
Total runs:
4.2K
Run Growth:
-191
Growth Rate:
-4.51%
Total runs:
3.2K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
3.0K
Run Growth:
-534
Growth Rate:
-18.05%
Total runs:
1.8K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
1.6K
Run Growth:
-134.7K
Growth Rate:
-8320.57%
Total runs:
1.6K
Run Growth:
1.6K
Growth Rate:
100.00%
Total runs:
1.6K
Run Growth:
421
Growth Rate:
26.80%
Total runs:
1.4K
Run Growth:
1.4K
Growth Rate:
100.00%
Total runs:
1.3K
Run Growth:
53
Growth Rate:
3.99%
Total runs:
1.2K
Run Growth:
1.2K
Growth Rate:
100.00%
Total runs:
1.1K
Run Growth:
-449
Growth Rate:
-40.89%
Total runs:
1.0K
Run Growth:
778
Growth Rate:
75.90%
Total runs:
894
Run Growth:
-2
Growth Rate:
-0.22%
Total runs:
855
Run Growth:
-3.0K
Growth Rate:
-356.02%
Total runs:
757
Run Growth:
516
Growth Rate:
68.16%
Total runs:
731
Run Growth:
731
Growth Rate:
100.00%
Total runs:
668
Run Growth:
668
Growth Rate:
100.00%
Total runs:
575
Run Growth:
575
Growth Rate:
100.00%
Total runs:
563
Run Growth:
-767
Growth Rate:
-136.23%
Total runs:
471
Run Growth:
-305
Growth Rate:
-64.76%
Total runs:
460
Run Growth:
-890
Growth Rate:
-193.48%
Total runs:
454
Run Growth:
0
Growth Rate:
0.00%
Total runs:
423
Run Growth:
133
Growth Rate:
31.44%
Total runs:
382
Run Growth:
34
Growth Rate:
8.90%
Total runs:
363
Run Growth:
69
Growth Rate:
19.01%
Total runs:
358
Run Growth:
128
Growth Rate:
35.75%
Total runs:
326
Run Growth:
0
Growth Rate:
0.00%
Total runs:
320
Run Growth:
286
Growth Rate:
89.38%
Total runs:
306
Run Growth:
121
Growth Rate:
39.54%
Total runs:
304
Run Growth:
5
Growth Rate:
1.64%
Total runs:
294
Run Growth:
59
Growth Rate:
20.07%
Total runs:
294
Run Growth:
-30
Growth Rate:
-10.20%
Total runs:
293
Run Growth:
2
Growth Rate:
0.68%
Total runs:
281
Run Growth:
119
Growth Rate:
42.35%
Total runs:
277
Run Growth:
70
Growth Rate:
25.27%
Total runs:
248
Run Growth:
0
Growth Rate:
0.00%
Total runs:
243
Run Growth:
62
Growth Rate:
25.51%
Total runs:
238
Run Growth:
161
Growth Rate:
67.65%
Total runs:
232
Run Growth:
-26
Growth Rate:
-11.21%
Total runs:
223
Run Growth:
73
Growth Rate:
32.74%
Total runs:
215
Run Growth:
102
Growth Rate:
47.44%
Total runs:
211
Run Growth:
-50
Growth Rate:
-23.70%
Total runs:
211
Run Growth:
0
Growth Rate:
0.00%
Total runs:
199
Run Growth:
76
Growth Rate:
38.19%
Total runs:
197
Run Growth:
53
Growth Rate:
26.90%
Total runs:
190
Run Growth:
51
Growth Rate:
26.84%
Total runs:
186
Run Growth:
31
Growth Rate:
16.67%
Total runs:
179
Run Growth:
-169
Growth Rate:
-94.41%
Total runs:
178
Run Growth:
110
Growth Rate:
61.80%
Total runs:
145
Run Growth:
58
Growth Rate:
40.00%
Total runs:
145
Run Growth:
0
Growth Rate:
0.00%
Total runs:
144
Run Growth:
120
Growth Rate:
83.33%
Total runs:
132
Run Growth:
-35
Growth Rate:
-26.52%
Total runs:
132
Run Growth:
132
Growth Rate:
100.00%
Total runs:
124
Run Growth:
0
Growth Rate:
0.00%
Total runs:
122
Run Growth:
50
Growth Rate:
40.98%
Total runs:
112
Run Growth:
20
Growth Rate:
17.86%
Total runs:
106
Run Growth:
44
Growth Rate:
41.51%
Total runs:
101
Run Growth:
0
Growth Rate:
0.00%
Total runs:
98
Run Growth:
50
Growth Rate:
51.02%
Total runs:
93
Run Growth:
13
Growth Rate:
14.13%
Total runs:
90
Run Growth:
11
Growth Rate:
12.36%
Total runs:
86
Run Growth:
1
Growth Rate:
1.16%
Total runs:
86
Run Growth:
39
Growth Rate:
45.35%
Total runs:
80
Run Growth:
-63
Growth Rate:
-78.75%