gpt54-mini-sequential-qwen3-4b-sft-s1-babyai-20260920
Qwen/Qwen3-4B trained with SFT using teacher gpt-5.4-mini. This is the final checkpoint of a completed babyai stage: five epochs, 125 optimizer updates in that stage.
Sequential stage 1 of BabyAI → TextCraft → SearchQA. Completed training prefix: babyai. Each environment receives five epochs. The student and method are carried through the chain.
Training-completion evidence: Observed complete stage manifest, verified full-step count, and complete controller task with verification marker.
No completed evaluation for this intermediate checkpoint is attached. Final three-environment scores belong only to the fully trained stage-3 model.
Provenance and interpretation notes:
- This is one trained checkpoint, not evidence of a general method advantage or replication across training seeds.
- The selection inventory records file names, sizes and modification times. It is not a tensor-byte hash binding to historical evaluation responses.
- No stage-specific evaluation is attached. Scores of later sequential checkpoints or standalone models must not be attributed to this checkpoint.
- This checkpoint is from a separate sequential chain, not the historical standalone run. Each stage carries the preceding stage of the same method and student.
The HF export is stored at the repository root; configuration, tokenizer, and all weight shards are included. Optimizer state, raw logs, and teacher trajectories are not included. Machine-readable source references and checksums are in experiment.json.
from transformers import AutoTokenizer, AutoModelForCausalLM
repo_id = "Diluner/gpt54-mini-sequential-qwen3-4b-sft-s1-babyai-20260920"
tokenizer = AutoTokenizer.from_pretrained(repo_id)
model = AutoModelForCausalLM.from_pretrained(repo_id, torch_dtype="auto")
Repository access is public, as authorized by the experiment owner. No license is asserted here; consult the base model and applicable terms.