58 lines
2.7 KiB
Markdown
58 lines
2.7 KiB
Markdown
|
---
|
||
|
|
license: other
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
library_name: transformers
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
tags:
|
||
|
|
- qwen3
|
||
|
|
- text-generation
|
||
|
|
- littlelearner
|
||
|
|
- bounded
|
||
|
|
- instruct
|
||
|
|
- reinforcement-learning
|
||
|
|
---
|
||
|
|
|
||
|
|
# littlelearner-0.6b-grpo-math-expert
|
||
|
|
|
||
|
|
0.617B K-5-bounded chat model post-trained with GRPO on top of SFT.
|
||
|
|
|
||
|
|
Part of the [**LittleLearner**](https://arxiv.org/abs/2608.13545) scale-up study (*pedagogically-controlled knowledge exposure*): Qwen3 dense LMs trained on a corpus filtered to U.S. K-5 material (**bounded**) vs an unfiltered FineWeb-Edu corpus (**unbounded**), to measure what an interpretable knowledge boundary costs and grants.
|
||
|
|
|
||
|
|
Note: This checkpoint was post-trained with GRPO on mathematical reasoning tasks to probe achievable performance on MathCAMPS. As a result, its behavior is specialized toward mathematical reasoning and may not preserve general-purpose chat capabilities; responses may also exhibit a tendency toward math-oriented reasoning or output.
|
||
|
|
|
||
|
|
## Model
|
||
|
|
- **Architecture:** Qwen3 dense (`Qwen3ForCausalLM`).
|
||
|
|
- **Size:** 0.617B params, hidden 1536, 20 layers, 12 query / 6 KV heads, FFN 4096. **Context:** 4096.
|
||
|
|
- **Tokenizer:** custom 64k byte-level BPE with per-digit splitting (ChatML special tokens).
|
||
|
|
- **Pretraining:** 88B tokens on K-5 **LittleCurriculum** (FineWeb-Edu filtered to U.S. grades K-5). WSD schedule, sharded Muon, MXFP8, Megatron-Core on 8xB200.
|
||
|
|
- **SFT:** supervised fine-tuned on K-5 chat data (lr 1e-5, 3 epochs) from a cooloff-SFT-primed base.
|
||
|
|
- **RL (GRPO):** segmented policy re-banding on a strictly K-5 verifiable-answer pool (Gemini-generated K-5 word problems + K-5-filtered GSM8K): 3 segments at rollout/training temperature 1.0, then 2 more at temperature 1.5 (the bounded-corpus, 0.6B-scale unlock temperature).
|
||
|
|
|
||
|
|
## Evaluation
|
||
|
|
MathCAMPS:
|
||
|
|
- K-5 pass@64 **61.1** / pass@1 **36.8**
|
||
|
|
- beyond-K-5 pass@64 **18.5** / pass@1 **6.4**
|
||
|
|
|
||
|
|
## Usage
|
||
|
|
```python
|
||
|
|
# transformers (chat)
|
||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
|
repo = "manueldeprada/littlelearner-0.6b-bounded-grpo"
|
||
|
|
tok = AutoTokenizer.from_pretrained(repo)
|
||
|
|
model = AutoModelForCausalLM.from_pretrained(repo, dtype="bfloat16", device_map="cuda")
|
||
|
|
msgs = [{"role": "user", "content": "Liam has 3 apples and buys 4 more. How many apples does he have?"}]
|
||
|
|
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)
|
||
|
|
out = model.generate(ids)
|
||
|
|
print(tok.decode(out[0, ids.shape[1]:], skip_special_tokens=True))
|
||
|
|
```
|
||
|
|
|
||
|
|
```python
|
||
|
|
# vLLM
|
||
|
|
from vllm import LLM
|
||
|
|
repo = "manueldeprada/littlelearner-0.6b-bounded-grpo"
|
||
|
|
llm = LLM(repo)
|
||
|
|
msgs = [{"role": "user", "content": "Liam has 3 apples and buys 4 more. How many apples does he have?"}]
|
||
|
|
print(llm.chat(msgs)[0].outputs[0].text)
|
||
|
|
```
|