-
Training Large Language Models To Reason In Parallel With Global Forking Tokens
Paper • 2510.05132 • Published • 1 -
shengjia-toronto/ssft32b_grpo_bs256_step10
Updated • 398 -
shengjia-toronto/ssft-32B-N6
Text Generation • 4B • Updated • 1.51k -
shengjia-toronto/grpo-test-ssft-32B
Text Generation • 33B • Updated • 17
Sheng Jia PRO
shengjia-toronto
·
AI & ML interests
None yet
Recent Activity
updated
a model
3 days ago
shengjia-toronto/grpo-test-ssft-32B
updated
a model
3 days ago
shengjia-toronto/ssft-32B-N6
updated
a collection
3 days ago
SSFT
Organizations
None yet
-
Training Large Language Models To Reason In Parallel With Global Forking Tokens
Paper • 2510.05132 • Published • 1 -
shengjia-toronto/ssft32b_grpo_bs256_step10
Updated • 398 -
shengjia-toronto/ssft-32B-N6
Text Generation • 4B • Updated • 1.51k -
shengjia-toronto/grpo-test-ssft-32B
Text Generation • 33B • Updated • 17
models
5
shengjia-toronto/grpo-test-ssft-32B
Text Generation
•
33B
•
Updated
•
17
shengjia-toronto/ssft-32B-N6
Text Generation
•
4B
•
Updated
•
1.51k
shengjia-toronto/ssft32b_grpo_bs256_step10
Updated
•
398
shengjia-toronto/ssft32b_grpo_bs256_step11_H200
33B
•
Updated
•
161
shengjia-toronto/sft_mixed_32b_thinktags_GAS4
4B
•
Updated
•
349
datasets
5
shengjia-toronto/ssft1Kcode-v2-Sonnet4-5-High-Run2-temp1-max29000
Viewer
•
Updated
•
1k
•
31
•
1
shengjia-toronto/ssft1Kcode-v2-Sonnet4-5-High-Run1-temp1-max29000
Viewer
•
Updated
•
1k
•
30
shengjia-toronto/ssft1Kcode-v2-OSS-High-Run1-temp1-max32768
Viewer
•
Updated
•
1k
•
23
shengjia-toronto/ssft1Kcode_v2
Viewer
•
Updated
•
1k
•
40
shengjia-toronto/dapo-math-17k-promptprocessed
Viewer
•
Updated
•
14.1k
•
46