mehuldamani/RLCR-hotpot-normal
Updated
Mehul Damani PRO
mehuldamani
AI & ML interests
Reinforcement Learning, Large Language Models
Recent Activity
published
a model about 17 hours ago
mehuldamani/sfted2_rlvr_multi__veryHardDataset_moreThinking published
a dataset 2 days ago
mehuldamani/multi-answer-sft-target-dataset published
a model 3 days ago
mehuldamani/sfted_rlvr_multi__veryHardDataset_moreThinking Organizations
None yet
mehuldamani/RLCR-hotpot-log-loss-v2
Updated
mehuldamani/llama-3.1-8b-base-user-sim-v1
Text Generation • 8B • Updated
• 25
mehuldamani/nov7_rerun_qwen3_8b_hotpot_rlcr_multiple
Updated
mehuldamani/nov7_rerun_qwen3_8b_hotpot_rlcr_single
Updated
mehuldamani/qwen3_8b_hotpot_rlvr_multiple_try2_nov7
Updated
mehuldamani/llama-3.1-8b-instruct-user-sim-v3
Text Generation • 8B • Updated
• 116
mehuldamani/llama-3.1-8b-instruct-user-sim-v2
Text Generation • 8B • Updated
• 1
mehuldamani/qwen3_8b_hotpot_rlVr_single_plzWork_nov7_1am
Updated
mehuldamani/qwen3_8b_hotpot_rlcr_single_plzWork_nov6AlmostMidnight
Updated
mehuldamani/qwen3_8b_hotpot_rlcr_multiple_try1
Updated
mehuldamani/qwen3_8b_hotpot_rlvr_single_oldParams
Updated
mehuldamani/qwen3_8b_hotpot_rlcr_single_oldParams
Updated
mehuldamani/qwen3_8b_hotpot_rlcr_single_1
Updated
mehuldamani/qwen3_8b_hotpot_rlvr_single
Updated
mehuldamani/qwen3_8b_hotpot_rlcr_single
Updated
mehuldamani/math-rlvr-v1-qwen3
Updated
mehuldamani/llama-3.1-8b-instruct-user-sim-sft-v3-global-step-600
Text Generation • 8B • Updated
• 1
mehuldamani/hotpot-qwen38b_nov1-rlcr-single-a100
Updated
mehuldamani/hotpot-qwen25b_nov4-rlvr-multiple-h100
Updated
mehuldamani/hotpot-sept27-rlvr-multiple-h100
Text Generation • 8B • Updated
• 1
mehuldamani/llama-3.1-8b-instruct-user-sim
Text Generation • 8B • Updated
• 6
mehuldamani/writing-qwen3-rl-v2
Updated
mehuldamani/writing-qwen3-rlav-v1
Updated
mehuldamani/debug-writing
Updated
mehuldamani/writing-qwen3-rl-v1
Updated
mehuldamani/writing-rlav-v2
Updated
mehuldamani/writing-rlav-v1
Updated
mehuldamani/writing-rl-v1
Updated
mehuldamani/hotpot-rlav-v2
Updated