namin/llm-verified-with-monte-carlo-tree-search issues and pull requests

#88 - run_reflexion generalized to support Llama3.1

Pull Request - State: closed - Opened by namin about 1 month ago

#87 - Refactor language into common scoring framework

Pull Request - State: closed - Opened by namin about 1 month ago - 1 comment

#86 - Stats

Pull Request - State: closed - Opened by namin about 2 months ago

#85 - adding RAG.md

Pull Request - State: closed - Opened by kemmaliu about 2 months ago

#84 - Issue with GPT-3.5/GPT-4 not continuing partial program as expected

Issue - State: open - Opened by 41968 about 2 months ago - 1 comment

#83 - new rag stuff, code gen still sometimes stuck

Pull Request - State: closed - Opened by kemmaliu about 2 months ago

#82 - Rag augment

Pull Request - State: closed - Opened by kemmaliu 3 months ago

#81 - Good Work! Any docs for code?

Issue - State: open - Opened by hsz0403 3 months ago - 2 comments

#80 - Python testing

Pull Request - State: closed - Opened by joaohts 3 months ago

#79 - Add plotting notebooks and final sweeps

Pull Request - State: closed - Opened by davidbrandfonbrener 3 months ago

#78 - docs for unittests

Pull Request - State: closed - Opened by shenniger 4 months ago

#77 - Python problems

Pull Request - State: closed - Opened by kemmaliu 4 months ago

#76 - Roadmap

Issue - State: open - Opened by namin 4 months ago

#75 - Unittests

Pull Request - State: closed - Opened by shenniger 4 months ago

#74 - add python language

Pull Request - State: closed - Opened by joaohts 4 months ago

#73 - changed image

Pull Request - State: closed - Opened by joaohts 4 months ago

#72 - added sandbox execution

Pull Request - State: closed - Opened by joaohts 4 months ago

#71 - add interleaving search

Pull Request - State: closed - Opened by shenniger 4 months ago

#70 - improve dafny script

Pull Request - State: closed - Opened by shenniger 4 months ago

#69 - add note about broken z3

Pull Request - State: closed - Opened by shenniger 4 months ago

#68 - more judged prompts

Pull Request - State: closed - Opened by shenniger 4 months ago

#67 - add more judged problems

Pull Request - State: closed - Opened by shenniger 4 months ago

#66 - Implement checks

Pull Request - State: closed - Opened by shenniger 4 months ago

#65 - added checks

Pull Request - State: closed - Opened by shenniger 5 months ago

#64 - Rollout and depth penalty

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#63 - add reflection sweep and wandb

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#62 - Reflexion var2

Pull Request - State: closed - Opened by sabrinahu5 5 months ago

#61 - Reflexion var1

Pull Request - State: closed - Opened by sabrinahu5 5 months ago

#60 - Sweeps

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#59 - fix node scorbale

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#58 - fixes for expansion

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#57 - Patch to handle EOS in limit_depth

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#56 - exploration sweep script

Pull Request - State: closed - Opened by shenniger 5 months ago

#55 - Reflexion

Pull Request - State: closed - Opened by shenniger 5 months ago

#54 - add workaround for StarCoder

Pull Request - State: closed - Opened by shenniger 5 months ago - 1 comment

#53 - Common wandb

Pull Request - State: closed - Opened by namin 5 months ago

#52 - add hparam notes

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#51 - add logging, add widen to root, reduce max depth penalty

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#50 - add logging of search stats

Pull Request - State: closed - Opened by davidbrandfonbrener 5 months ago

#49 - ready for sweeps

Pull Request - State: closed - Opened by shenniger 5 months ago

#48 - add more wandb logging and a draft slurm script

Pull Request - State: closed - Opened by shenniger 6 months ago

#47 - Add wandb to run_intermediate_expansion for simon

Pull Request - State: closed - Opened by davidbrandfonbrener 6 months ago

#46 - add wandb for run_whole and fix a few things with token_counter

Pull Request - State: closed - Opened by davidbrandfonbrener 6 months ago

#45 - Fix token count

Pull Request - State: closed - Opened by shenniger 6 months ago

#44 - Implement token limit

Pull Request - State: closed - Opened by shenniger 6 months ago

#43 - implemented node duplication, so far only for logging

Pull Request - State: closed - Opened by shenniger 6 months ago - 1 comment

#42 - Crl/run whole on dafny

Pull Request - State: closed - Opened by ChloeL19 7 months ago

#41 - Crl/clover parametric

Pull Request - State: closed - Opened by ChloeL19 7 months ago

#40 - Crl/whole sampling

Pull Request - State: closed - Opened by ChloeL19 7 months ago

#39 - Crl/intermediate expansion

Pull Request - State: closed - Opened by ChloeL19 7 months ago

#38 - Crl/add clover

Pull Request - State: closed - Opened by ChloeL19 7 months ago

#37 - moved extra language about huggingface in readme to comment

Pull Request - State: closed - Opened by ChloeL19 8 months ago

#36 - Added instructions to README about running the DPO script on only one GPU.

Pull Request - State: closed - Opened by ChloeL19 8 months ago

#35 - in the README setup I add a quick note that developers need to login to huggingface to use the models there before running python run.py

Pull Request - State: closed - Opened by ChloeL19 8 months ago

#34 - in execute.py, I ensure the file is opened with the encoding utf-8. This fixed a bug I had on my end.

Pull Request - State: closed - Opened by ChloeL19 8 months ago

#33 - Re-ran some results

Pull Request - State: closed - Opened by sibiraja 8 months ago

#32 - Redo experiment logs

Pull Request - State: closed - Opened by sibiraja 8 months ago

#31 - Some more results

Pull Request - State: closed - Opened by sibiraja 8 months ago

#30 - Added stats for timeouts

Pull Request - State: closed - Opened by sibiraja 8 months ago

#29 - Multiple experiments pipeline

Pull Request - State: closed - Opened by sibiraja 8 months ago

#28 - Factorial experiments

Pull Request - State: closed - Opened by namin 9 months ago

#27 - initial graph code

Pull Request - State: closed - Opened by shenniger 9 months ago

#26 - Added bash script to log cli flags and time experiments

Pull Request - State: closed - Opened by sibiraja 9 months ago

#25 - Edited readme to reflect cli usage and removed cli test files

Pull Request - State: closed - Opened by sibiraja 9 months ago

#24 - Added command line flags to codebase

Pull Request - State: closed - Opened by sibiraja 9 months ago - 1 comment

#23 - added two prompts for parsers

Pull Request - State: closed - Opened by shenniger 9 months ago

#22 - Meta

Pull Request - State: closed - Opened by namin 9 months ago

#21 - fix bug in feature indexing

Pull Request - State: closed - Opened by davidbrandfonbrener 10 months ago

#20 - add run_diversity, return hiddens from llm

Pull Request - State: closed - Opened by davidbrandfonbrener 10 months ago - 1 comment

#19 - Focus

Pull Request - State: closed - Opened by namin 10 months ago

#18 - fix: huggingface typo

Pull Request - State: closed - Opened by johnjyang 10 months ago

#17 - fix: huggingface typo

Pull Request - State: closed - Opened by johnjyang 10 months ago

#16 - Refactor and add preliminary types

Pull Request - State: closed - Opened by johnjyang 10 months ago - 8 comments

#15 - fix: timeout error

Pull Request - State: closed - Opened by johnjyang 10 months ago

#14 - Improve lean4 prompt

Pull Request - State: closed - Opened by johnjyang 10 months ago - 1 comment

#13 - Bug in Lean checker

Issue - State: closed - Opened by namin 10 months ago - 4 comments

#12 - Add Mathlib support

Pull Request - State: closed - Opened by johnjyang 10 months ago

#11 - extended max_completion_depth logic to all run_* files

Pull Request - State: closed - Opened by sibiraja 10 months ago

#10 - integrated max_completion_depth limit for recursion limit

Pull Request - State: closed - Opened by sibiraja 10 months ago

#9 - port @sibiraja max_depth to all runs where it makes sense

Pull Request - State: closed - Opened by namin 10 months ago

#8 - integrated max_depth feature

Pull Request - State: closed - Opened by sibiraja 10 months ago - 1 comment

#7 - Use local execution of verifiers

Pull Request - State: closed - Opened by namin 10 months ago - 1 comment

#6 - Lean tweaks

Pull Request - State: closed - Opened by namin 10 months ago - 1 comment

#5 - Add support for Lean

Pull Request - State: closed - Opened by johnjyang 10 months ago - 10 comments

#4 - Verifier feedback

Pull Request - State: closed - Opened by namin 10 months ago

#3 - Ask user

Pull Request - State: closed - Opened by namin 10 months ago

#2 - this is cool, thanks.

Issue - State: closed - Opened by russellballestrini 10 months ago

#1 - PPO

Pull Request - State: closed - Opened by namin 10 months ago

GitHub / namin/llm-verified-with-monte-carlo-tree-search issues and pull requests