reproducing DeepSeek R1 Zero with Qwen2.5-0.5B on two 4090 GPUs
rasdani PRO
rasdani
AI & ML interests
None yet
Recent Activity
liked
a dataset
7 days ago
R2E-Gym/R2EGym-TestingAgent-SFT-Trajectories
published
a dataset
9 days ago
rasdani/SkyRL-v0-293-data-oracle-4k-context-100-epochs
liked
a model
9 days ago
StringChaos/R2E-TestgenAgent
Organizations
models
37
rasdani/deepseek_r1_qwen14b_swe_rl_8k
15B
•
Updated
•
5
rasdani/deepseek_r1_llama_8b_swe_rl_8k_12_epochs
8B
•
Updated
•
6
rasdani/qwen3_8b_swe_rl_8k
8B
•
Updated
•
18
rasdani/deepseek_r1_7b_gh_patches_2k_fixed_reward
8B
•
Updated
•
7
rasdani/deepseek_r1_7b_gh_patches_2k
8B
•
Updated
•
4
rasdani/crux-eval_math-eval-logs
Updated
rasdani/git-diff-Qwen-4B-10k
4B
•
Updated
•
3
rasdani/git-diff-Qwen-4B-10k-checkpoints
Updated
rasdani/git-diff-Qwen-4B-32k-checkpoints
Updated
rasdani/git-diff-Qwen-4B-30k
4B
•
Updated
•
4
datasets
169
rasdani/SkyRL-v0-293-data-oracle-4k-context-100-epochs
Viewer
•
Updated
•
7.2k
•
76
rasdani/SkyRL-v0-293-data-oracle-8k-context-100-epochs
Viewer
•
Updated
•
13.1k
•
96
rasdani/deepseek_r1_llama_8b_swe_rl_8k_12_epochs_preds_100_v2
Viewer
•
Updated
•
100
•
81
rasdani/DeepSeek-R1-Distill-Llama-8B_preds_100
Viewer
•
Updated
•
100
•
80
rasdani/SWE-bench_Verified_oracle-parsed_commits_32k_100
Viewer
•
Updated
•
100
•
98
rasdani/SWE-bench_oracle-parsed-commits_100
Viewer
•
Updated
•
100
•
85
rasdani/SWE-bench_oracle-parsed-commits
Viewer
•
Updated
•
2.29k
•
90
rasdani/deepseek_r1_qwen14b_swe_rl_8k_results
Updated
•
33
rasdani/deepseek_r1_qwen14b_swe_rl_8k_56_steps_preds
Viewer
•
Updated
•
465
•
76
rasdani/deepseek_r1_qwen14b_swe_rl_8k_56_steps_preds_100
Viewer
•
Updated
•
100
•
46