reproducing DeepSeek R1 Zero with Qwen2.5-0.5B on two 4090 GPUs
rasdani
rasdani
AI & ML interests
None yet
Recent Activity
liked
a dataset
1 day ago
Skywork/Skywork-Reward-Preference-80K-v0.2
liked
a model
1 day ago
mistralai/Devstral-Small-2505
new activity
2 days ago
reyavir/PromptEvals:Report 200 criteria which were manually validated
Organizations
Collections
1
Papers
1
models
23
rasdani/qwen3_0_6b_function_rm
Updated
•
14
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-8192k
Updated
•
1
rasdani/Qwen2.5-0.5B-simpleRL-Zoo
Text Generation
•
Updated
•
5
rasdani/smolR1-Qwen2.5-0.5B
Text Generation
•
Updated
•
6
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-no-KL
Updated
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-3072k
Updated
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-4096k
Updated
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-2560k
Updated
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-2048k
Updated
rasdani/Qwen2.5-0.5B-simpleRL-Zoo-first-try
Updated
•
2
datasets
98
rasdani/swe-fixer-70k-filtered
Viewer
•
Updated
•
54.3k
•
50
rasdani/ifeval-genesys
Viewer
•
Updated
•
15k
•
41
rasdani/ifeval-genesys-verified
Viewer
•
Updated
•
48
•
38
rasdani/ifeval-genesys-debug
Viewer
•
Updated
•
48
•
41
rasdani/swe-fixer-debug-DeepSeek-R1-verified
Viewer
•
Updated
•
30
•
170
rasdani/swe-fixer-debug-DeepSeek-R1
Viewer
•
Updated
•
30
•
78
rasdani/swe-fixer-70k
Viewer
•
Updated
•
69.8k
•
147
rasdani/swe-fixer-debug-Qwen3-235B-A22B-verified
Viewer
•
Updated
•
10
•
32
rasdani/swe-fixer-debug-pi-format
Viewer
•
Updated
•
100
•
60
rasdani/reasoning-gym-dataset-debug-pi-format
Viewer
•
Updated
•
20k
•
12