Downloads · 30 days
10
9% of all-time downloads
laion/a3-rl-laion_nemotron-gym-agent-workplace-v2-5-8B
a3-rl-laion_nemotron-gym-agent-workplace-v2-5-8B is a machine learning model from laion. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
RL (GRPO / rloon) checkpoint, globalstep5, trained from laion/GLM-47-swesmith-sandboxes-withtests-oracleverified120s-maxeps-131k-fixthink (Qwen3-8B architecture) on open-athena/nemotron-gym-agent-workplace-v2.
Downloads · 30 days
10
9% of all-time downloads
All-time downloads
111
Public
Parameters
8.2B
16.4 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors16.4 GB · 100%
From the Hugging Face model README
RL (GRPO / rloo_n) checkpoint, global_step_5, trained from
laion/GLM-4_7-swesmith-sandboxes-with_tests-oracle_verified_120s-maxeps-131k-fixthink
(Qwen3-8B architecture) on open-athena/nemotron-gym-agent-workplace-v2.
a3 series #21 — small, data-limited run. The dataset is only 297 tasks, so
2 epochs exhausted at ~step 9 (NOT the max_steps=80 budget). Checkpoint
global_step_5 was selected by the highest 5-period EMA (alpha=1/3) of
reward/avg_raw_reward (EMA@5 = 0.3987, raw reward@5 = 0.4746). The step-9
export emitted no logged reward and the trailing trajectory was flat/declining
(s1..s8: 0.541, 0.479, 0.287, 0.197, 0.475, 0.355, 0.357, 0.252). This is a
legitimate but weak a3 ablation data point, not a strong checkpoint.
Training-time Daytona/Harbor rollouts for this run are uploaded as a companion dataset: open-athena/a3-rl-laion_nemotron-gym-agent-workplace-v2
The dataset contains the last episode of each trial (per
make_and_upload_trace_dataset --episodes last) — the same rollouts
the policy was trained on after rollback / truncation.