Modelpublic

AdithyaSK/data-agent-2b-normal-final

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes17downloads

Model Card

data-agent-2b-normal-final (v0)

A 2B data-science agent finetuned from `Qwen/Qwen3.5-2B` with GRPO (online RL) to solve data-analysis tasks in a sandboxed bash environment. This repo holds the final checkpoint (end of a full 1-epoch run) of the 2b-normal run.

Training

—Method: GRPO (Group Relative Policy Optimization) via TRL.
—Environment: Harbor task spec + E2B cloud sandboxes; single bash tool, answer submitted to /workdir/answer.txt.
—Dataset: `AdithyaSK/data_agent_rl_environment_train`.
—Schedule: 1 epoch (1119 steps), 8 generations/prompt, KL-anchored to the reference. Tasks were presented in random order (standard).
—This checkpoint: step 1119 (final).

Evaluation

Agentic pass@k on the held-out `data_agent_rl_environment_eval` suite (366 tasks, 4 samples/task, unbiased estimator):

metric	base (Qwen3.5-2B)	this model	Δ
pass@1	0.098	0.408	+0.310
pass@2	0.168	0.514	+0.346
pass@3	0.229	0.567	+0.338
pass@4	0.284	0.603	+0.319

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer
m = AutoModelForCausalLM.from_pretrained("AdithyaSK/data-agent-2b-normal-final", revision="v0", torch_dtype="bfloat16")
tok = AutoTokenizer.from_pretrained("AdithyaSK/data-agent-2b-normal-final", revision="v0")

Part of the data-agent v0 release. Served non-thinking with a single `bash` tool (Qwen tool-calling).