kshitijthakkar/repro-differentiable-conformal-training-artifacts
90.4 kB
Kshitij Thakkar PRO
kshitijthakkar
AI & ML interests
Building the evaluation and observability layer for AI.
Creator of TraceVerse—turning real-world LLM interactions into datasets, benchmarks, and cost-efficient model insights.
Recent Activity
updated a dataset 1 day ago
kshitijthakkar/smoltrace-leaderboard updated a Space 1 day ago
kshitijthakkar/repro-online-inverse-linear-opt updated a bucket 1 day ago
kshitijthakkar/repro-online-inverse-linear-opt-artifactsOrganizations
Articles 6
Article
The Mind of Tashi: making a 200M-active model's reasoning *the game*
Article
2
Scaling Mixture of Experts: Architecture Search for Billion-Parameter Language Models
Trackio logbooks: claim-by-claim reproductions of ICML 2026 papers for the Agent Reproducibility Challenge.
-
Repro - MemEvolve: Meta-Evolution of Agent Memory Systems
🧬Collaborate with an AI agent to manage a shared experiment logbook
-
Repro - TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
🎯Explore experiment logs and sync findings with a coding agent
-
Repro - How to Correctly Report LLM-as-a-Judge Evaluations
🎯Log and share LLM evaluation findings in a collaborative notebook
-
Repro - Dependence-Aware Label Aggregation via Ising Models
🧲Explore and edit experiment logbooks with AI agent help
Qwen3.6-35B-A3B carved by expert importance to fit one consumer GPU. No training, no calibration. 800 tok/s on a laptop 5090.
-
kshitijthakkar/Kirigami-Qwen3.6-20B-A3B-NVFP4
Text Generation • 14B • Updated • 70 • 1 -
kshitijthakkar/Kirigami-Qwen3.6-24B-A3B-NVFP4
Text Generation • 16B • Updated • 276 -
kshitijthakkar/Kirigami-Qwen3.6-28B-A3B-NVFP4
Text Generation • 19B • Updated • 20 • 1 -
Kirigami Journey
🪷How we carved a 35B MoE to fit a 24GB GPU — zero training
Trackio logbooks: claim-by-claim reproductions of ICML 2026 papers for the Agent Reproducibility Challenge.
-
Repro - MemEvolve: Meta-Evolution of Agent Memory Systems
🧬Collaborate with an AI agent to manage a shared experiment logbook
-
Repro - TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
🎯Explore experiment logs and sync findings with a coding agent
-
Repro - How to Correctly Report LLM-as-a-Judge Evaluations
🎯Log and share LLM evaluation findings in a collaborative notebook
-
Repro - Dependence-Aware Label Aggregation via Ising Models
🧲Explore and edit experiment logbooks with AI agent help
Qwen3.6-35B-A3B carved by expert importance to fit one consumer GPU. No training, no calibration. 800 tok/s on a laptop 5090.
-
kshitijthakkar/Kirigami-Qwen3.6-20B-A3B-NVFP4
Text Generation • 14B • Updated • 70 • 1 -
kshitijthakkar/Kirigami-Qwen3.6-24B-A3B-NVFP4
Text Generation • 16B • Updated • 276 -
kshitijthakkar/Kirigami-Qwen3.6-28B-A3B-NVFP4
Text Generation • 19B • Updated • 20 • 1 -
Kirigami Journey
🪷How we carved a 35B MoE to fit a 24GB GPU — zero training
spaces 41
pinned
Sleeping
Agents
GuardianTails
🐾
Pet Health Intelligence Platform
Running
2604.20098
📐
Explore and collaborate on your project logbook
Running
Repro - Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets
📐
Explore and collaborate on a research logbook online
Running
Repro - TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
🎯
Explore experiment logs and sync findings with a coding agent
Running
Repro - Gaussian Mean Field Variational Inference can Overestimate Predictive Variance
📉
Explore research logbook and sync with your coding agent
Running
Reproducing GRPO's Loss, Dynamics, and Success Amplification (arXiv:2503.06639)
📈
Browse a research logbook and collaborate with an AI agent
models 149
kshitijthakkar/loggenix-moe-1.3B-base-cpt-test
Updated
kshitijthakkar/Kirigami-Qwen3.6-28B-A3B-NVFP4
Text Generation • 19B • Updated • 20 • 1
kshitijthakkar/Kirigami-Qwen3.6-20B-A3B-NVFP4
Text Generation • 14B • Updated • 70 • 1
kshitijthakkar/Kirigami-Qwen3.6-24B-A3B-NVFP4
Text Generation • 16B • Updated • 276
kshitijthakkar/deepseek-v4-mini-300M-sft-v2
Text Generation • 0.3B • Updated • 313
kshitijthakkar/deepseek-v4-mini-300M-sft-v1
Text Generation • 0.3B • Updated • 303
kshitijthakkar/deepseek-v4-mini-300M-cpt-fineweb
Text Generation • 0.3B • Updated • 304
kshitijthakkar/deepseek-v4-mini-300M-recovered
Text Generation • 0.3B • Updated • 33 • 1
kshitijthakkar/deepseek-v4-mini-300M-recovered-h100
Text Generation • 0.3B • Updated • 5
kshitijthakkar/deepseek-v4-mini-300M-recovered-wip
Text Generation • 0.3B • Updated • 6
datasets 423
kshitijthakkar/smoltrace-leaderboard
Viewer • Updated • 111 • 59
kshitijthakkar/Open-SWE-Traces
Viewer • Updated • 207k • 135
kshitijthakkar/EEE_datastore
Viewer • Updated • 4.89k • 244
kshitijthakkar/Nemotron-Personas-India
Viewer • Updated • 3M • 218
kshitijthakkar/AgentTrove
Viewer • Updated • 1.7M • 45
kshitijthakkar/GLM-5.1-Reasoning-1M-Cleaned
Preview • Updated • 113
kshitijthakkar/smoltrace-metrics-20260424_122614
Viewer • Updated • 1 • 15
kshitijthakkar/smoltrace-traces-20260424_122614
Viewer • Updated • 2 • 13
kshitijthakkar/smoltrace-results-20260424_122614
Viewer • Updated • 2 • 13
kshitijthakkar/smoltrace-metrics-20260424_112312
Viewer • Updated • 1 • 10