Dataset Viewer

The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.

ALFWorld OLMo-3-7B-Instruct adaptive-pivot TERSE: held-out evaluation

Held-out ALFWorld evaluation rows for wckwan/Alfworld-Olmo3-7B-Adaptive-Pivot, the terse-selector control for the ALFWorld OLMo-3 pivot cell. That training run crashed at step 77, so only gs20, gs40 and gs60 exist.

Pairs with the FENCED arm in wckwan/Alfworld-Olmo3-7B-Adaptive-Pivot-Fenced-eval, evaluated on the same box, tree and protocol, for a terse-vs-fenced contrast at gs20-60.

Layout

global_step_N/result.json         # metrics, diagnostics, task list, config, fingerprint, runtime
global_step_N/trajectories.jsonl  # per-episode trajectories
global_step_N/eval_seed0.log      # eval log
MANIFEST.sha256                   # sha256sum -c compatible

Protocol

standalone_eval preset alfworld_olmo3: eval_in_distribution, env_seed 0, 128 goals, max_steps 50, prompt/response 2048/2048, history_length 2, temperature 0.4, top_p 1.0, top_k -1, chat kwargs {}, history_action_char_cap None. Runtime: vLLM 0.11.0, data parallel 2 (TP=1), engine seed 0; verl-agent tree 8abb4e2d (includes the final-</think> action parser 451b4969). Every row passes scripts/check_result_alfworld.py (gamefiles hash 12ead9df9760a8a8).

With 128 goals, the sampling standard deviation of a success rate is about 4.4 points. Treat differences under about 9 points as not separated.

Config fingerprint

fingerprint
these rows (native) 942c1e75645e9f07f6ecd91a66d3ddbf1b34a87097623862db0467e06bca7fcc
baseline rows / legacy-equivalent c12b948ddb980df68520982077f8202cc260b1abd21f9194e06c4eae7012abc3

The only difference is the history_action_char_cap key (commit a668c6b6), None here and behaviourally a no-op; removing it reproduces the baseline hash exactly.

Downloads last month
37