erv1n/e2l-alfworld-qwen3-8b-coldstart
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
c140513f129b645465972641e6c4ed6330020399
e2l-alfworld-qwen3-8b-colds…/README.md
ModelHub XC c140513f12 初始化项目，由ModelHub XC社区提供模型 
Model: erv1n/e2l-alfworld-qwen3-8b-coldstart
Source: Original Platform
2026-09-18 16:54:23 +08:00

811 B
Raw Blame History

library_name, license
library_name license
transformers apache-2.0

E2L ALFWorld cold-start checkpoint (Qwen/Qwen3-8B)

Stage-1 cold-start checkpoint from Ch1nyzzz/e2-learning: full-parameter online mistake-driven experience learning (next-observation SFT gated by a semantic judge) on ALFWorld AlfredTWEnv, initialized from Qwen/Qwen3-8B. Source run artifact: outputs/alfworld_qwen3_8b_active_10000_hf/checkpoints/env_step_006220.

Intended use: initialization ("cold start") for Stage-2 policy GRPO training with verl-agent. Load as a regular Hugging Face checkpoint (AutoModelForCausalLM.from_pretrained). controller_state.json is experiment-controller provenance metadata and can be ignored by inference/training stacks.