salmannyu/llama_base_thinking_sft_noisy_reward_0_9
Updated
Salman Rahman PRO
salmannyu
AI & ML interests
Natural Language Processing, Deep Learning, Scalable Oversight, and Language Model Evaluation
Recent Activity
upvoted a paper 4 days ago
When Can LLMs Learn to Reason with Weak Supervision? submitted a paper 4 days ago
When Can LLMs Learn to Reason with Weak Supervision? updated a collection 4 days ago
rlvr-weak-supervisionOrganizations
salmannyu/llama_base_thinking_sft_majority_vote_math_1024_sample_8k
Updated
salmannyu/mid_train_llama_52b_thinking_data_effect_math_8_sample
Updated
salmannyu/mid_train_llama_52b_thinking_noisy_reward_math_0.7_sample
Updated
salmannyu/mid_train_llama_52b_thinking_noisy_reward_math_0.9_sample
Updated
salmannyu/mid_train_llama_52b_thinking_majority_vote_math_1024_sample
Updated
salmannyu/mid_train_llama_52b_thinking_data_effect_math_2048_sample
Updated
salmannyu/data_effect_scp_do_llama_3b_2048_sample
Updated
salmannyu/data_effect_scp_do_llama_3b_8_sample
Updated
salmannyu/data_effect_math_do_llama_3b_8_sample
Updated
salmannyu/data_effect_math_do_qwen_1_5b_8_sample
Updated
salmannyu/Llama-3B-Nemotron-mid-think_sft_nopack_lr1.5e5_ep3
3B • Updated • 1
salmannyu/Llama-3B-Nemotron-Math-thinking-sft-3ep-8samp-default-step150
4B • Updated • 2
salmannyu/Llama-3B-Nemotron-Math-thinking-sft-3ep-8samp-default-step100
4B • Updated • 1
salmannyu/Llama-3B-Nemotron-Math-Mid-Train-Full-non-think-nopack-lr1.5e5-ep3
3B • Updated • 84
salmannyu/Llama-3B-Nemotron-Math-Mid-Train-Full-nopack-lr1.5e5-ep3
3B • Updated • 1
salmannyu/Llama-3B-Nemotron-Math-Mid-Train-Full
Text Generation • 3B • Updated • 2
salmannyu/Llama-3B-Nemotron-Math-Mid-Train-140K-Step
3B • Updated • 1
salmannyu/Qwen2.5-1.5B-Nemotron-Math-52B-Mid-Train-8
Text Generation • 2B • Updated • 4
salmannyu/nemotron-train8-52B-Token
2B • Updated • 2
salmannyu/nemotron-train4
2B • Updated • 2
salmannyu/train3
2B • Updated
salmannyu/nemotron-train2
2B • Updated • 1