[D-27] Line Completion - / ing
📝 논문 Literature Review 준비 새 채팅에서 다음을 준비하면 좋겠습니다: 필요한 정보: 논문 제목: “Empirical Study of Autoregressive State Space Models for Software Development” 주요 섹션: Related Work (SSMs, Code Models) Background ...
📝 논문 Literature Review 준비 새 채팅에서 다음을 준비하면 좋겠습니다: 필요한 정보: 논문 제목: “Empirical Study of Autoregressive State Space Models for Software Development” 주요 섹션: Related Work (SSMs, Code Models) Background ...
오늘 미팅 내용이 아주 좋지 않았다…잡도리…
#Token-level Fine-tuning <==> Raw results; <BiGS> (_) [kim@slurm-login causal]$ cat token_completion_py150/test_results.json { "eval_loss": 1.1523972749710083, "eval_accuracy": 0.7...
📋 작업 요약 (2026-01-20) 🎯 목표 BiGS와 Mamba 모델을 CodeXGLUE line completion 태스크로 fine-tuning 🔴 발견된 문제들 Dataset 문제 (Root Cause) code_x_glue_cc_code_completion_line 데이터셋의 gt 필드가 완전히 비어있음 결과: loss=0.0, eval...
line completion token completion and then code generation https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/CodeCompletion-line https://github.com/ishuoliu/ASE-CMamba/tree/main
1. 기존 mamba 환경 삭제 (선택사항) conda deactivate conda env remove -n mamba 2. 새 환경 생성 (Python 3.10 권장) conda create -n mamba python=3.10 -y conda activate mamba 3. GCC 9+ 설치 (conda로) conda install -c co...
w and b perplexity and plots increasing the number of lambda and max span length. next token prediction next token prediction
Project Sync / Status Update Summary Model Configuration Issues Vocabulary Size Problem: The current configuration has an incorrect vocab size (32101 instead of 32102) causing index erro...
set up /kim/causal/modeling_bigs_gen.py /kim/causal/configuration_bigs.py /kim/causal/train_fim.py /shared/STG/zhang_dataset/issues_eval.jsonl issues_train.jsonl...
Remote-SSH Connection Instability Report This issue describes a failure in establishing a stable SSH connection from VS Code using the Remote-SSH extension, leading to repeated disconnections an...