Article
Hajime Yagihara
HachiML
AI & ML interests
LLM, Synthetic data
Recent Activity
published an article about 5 hours ago
Decision Models: Their Limits and Reinforcement Learning updated a Space over 1 year ago
HachiML/simple_data_analyst published a Space over 1 year ago
HachiML/simple_data_analystOrganizations
Reasoning Vector
Reasoningモデルとベースモデルの重み差分
-
HachiML/Llama-3.1-Swallow-8B-v0.2-reasoningvector-deepseek-r1
Text Generation • 8B • Updated • 34 • 2 -
HachiML/ReasoningVector-DeepSeek-R1-Distill-Llama-8B
8B • Updated • 26 -
HachiML/ReasoningVector-Mistral-Small-24B-Instruct-2501-reasoning
24B • Updated • 6 -
HachiML/ReasoningVector-DeepSeek-R1-Distill-Qwen-32B
33B • Updated • 19
Reasoning Vector
Reasoningモデルとベースモデルの重み差分
-
HachiML/Llama-3.1-Swallow-8B-v0.2-reasoningvector-deepseek-r1
Text Generation • 8B • Updated • 34 • 2 -
HachiML/ReasoningVector-DeepSeek-R1-Distill-Llama-8B
8B • Updated • 26 -
HachiML/ReasoningVector-Mistral-Small-24B-Instruct-2501-reasoning
24B • Updated • 6 -
HachiML/ReasoningVector-DeepSeek-R1-Distill-Qwen-32B
33B • Updated • 19
Reasoning Model Evaluation Results
OpenAI o1再現の取り組み
models 161
HachiML/ReasoningVector-DeepSeek-R1-Distill-Qwen-32B
33B • Updated • 19
HachiML/ReasoningVector-Mistral-Small-24B-Instruct-2501-reasoning
24B • Updated • 6
HachiML/Llama-3.1-Swallow-8B-v0.2-reasoningvector-deepseek-r1
Text Generation • 8B • Updated • 34 • 2
HachiML/ReasoningVector-DeepSeek-R1-Distill-Llama-8B
8B • Updated • 26
HachiML/QwQ-CoT-0.5B-JA-v1.1
Text Generation • 0.5B • Updated • 21
HachiML/QwQ-CoT-0.5B-JA-v1.0
Text Generation • 0.5B • Updated • 13
HachiML/QwQ-CoT-0.5B-JA
Text Generation • 0.5B • Updated • 24
HachiML/QwQ-CoT-0.5B-JA-v0.9
Text Generation • 0.5B • Updated • 11
HachiML/QwQ-CoT-0.5B-JA-v0.8
Text Generation • 0.5B • Updated • 16
HachiML/QwQ-CoT-0.5B-JA-v0.7
Text Generation • 0.5B • Updated • 13
datasets 77
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1-MCTS-ips5-mi15-mss64-et0-3
Viewer • Updated • 250 • 4
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1-MCTS-ips13-mi15-mss32-et0-sa
Viewer • Updated • 250 • 7
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1-MCTS-ips15-mi15-mss64-et0-sa
Viewer • Updated • 250 • 4
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1-MCTS-ips12-mi15-mss32-et0-sa
Viewer • Updated • 250 • 5
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1-MCTS-ips8-mi15-mss32-et0-sa
Viewer • Updated • 250 • 14
HachiML/mgsm_250-QwQ-MCTS-0.5B-JA-v1.1-ips5-mi15-mss64-et0-sa
Viewer • Updated • 250 • 4
HachiML/mgsm_250-QwQ-MCTS-0.5B-JA-v1.1-ips5-mi15-mss32-et0-sa
Viewer • Updated • 250 • 5
HachiML/mgsm_250-QwQ-MCTS-0.5B-JA-v1.1-ips5-mi15-mss32-et0
Viewer • Updated • 250 • 10
HachiML/mgsm_250-QwQ-MCTS-0.5B-JA-v1.1-MCTS-ips5-mi15-mss64-et0
Viewer • Updated • 250 • 10
HachiML/mgsm_250-QwQ-CoT-0.5B-JA-v1.1
Viewer • Updated • 250 • 8