Downloads · 30 days
15
47% of all-time downloads
budget-internalization-iclr2027
gemma4-e2b-2k-grpo-superworm-s300
budget-internalization-iclr2027/gemma4-e2b-2k-grpo-superworm-s300
gemma4-e2b-2k-grpo-superworm-s300 is a any-to-any model from budget-internalization-iclr2027. Use it for the any-to-any task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.
RL-finetuned google/gemma-4-E2B-it trained for math reasoning under a 2,048-token generation budget. Released as part of an anonymous ICLR 2027 submission.
All-time downloads
32
Public
Parameters
5.1B
20.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors20.5 GB · 100%
At a glance
- Task
- Any-to-Any
- Library
- transformers
- License
- apache-2.0
- Model type
- gemma4
- Access
- Public
- Created
- Sep 23, 2026
- Updated
- Sep 23, 2026
- SHA
- 413f5a58
Base models
Model card
From the Hugging Face model README
Gemma-4-E2B-it · 2k-token budget · grpo · superworm
RL-finetuned google/gemma-4-E2B-it trained for math reasoning under a 2,048-token generation budget. Released as part of an anonymous ICLR 2027 submission.
Run ID (petname): superworm · checkpoint step 300
Training
| Base model | google/gemma-4-E2B-it |
| Algorithm | GRPO (leave-one-out baseline, group reward normalization, token-level loss). Responses that hit the budget are truncated and scored as-is. |
Generation budget (max_new_tokens) | 2,048 |
| Data | DeepScaleR (math), 3 epochs max |
| Batch | 32 prompts × 8 rollouts per step |
| Optimizer | Adam, cosine LR schedule, peak LR 3e-6, 10 warmup steps |
| Steps | 300 |
| Reward | binary answer correctness (\boxed{} extraction) |
| Weights dtype | F32 |
Training prompt (user turn, rendered with the base model's chat template):
Think step-by-step to solve the following problem. Output your answer inside of \\boxed{} tags.:
{problem}
Let's think step-by-step
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "budget-internalization-iclr2027/gemma4-e2b-2k-grpo-superworm-s300"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype="auto", device_map="auto")
With vLLM: vllm serve budget-internalization-iclr2027/gemma4-e2b-2k-grpo-superworm-s300
License
Inherits the license of the base model (google/gemma-4-E2B-it).
- Task
- Any-to-Any
- Library
- transformers
- Type
- gemma4
- License
- apache-2.0
- Created
- Sep 23, 2026
- Updated
- Sep 23, 2026