Upload README.md with huggingface_hub

README.md

@@ -1,202 +1,110 @@
 ---
-base_model: cmpatino/qwen-grpo-r4-s100
+license: apache-2.0
+base_model: Qwen/Qwen3-0.6B
+datasets:
+- openai/gsm8k
+library_name: transformers
+pipeline_tag: text-generation
 tags:
+- trl
+- grpo
 - trl-autoresearch
 - qwen-grpo
-- grpo
+- gsm8k
+- reasoning
+model-index:
+- name: qwen-grpo-r5
+  results:
+  - task:
+      type: text-generation
+      name: Math word problems
+    dataset:
+      name: GSM8K (test)
+      type: openai/gsm8k
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.7089
+      name: accuracy (inspect_evals/gsm8k, 10-shot, greedy)
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
-
-## Model Details
-
-### Model Description
-
-<!-- Provide a longer summary of what this model is. -->
-
-
-
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-
-### Model Sources [optional]
-
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
-
-### Recommendations
-
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
+# Qwen3-0.6B + GRPO on GSM8K
 
-### Compute Infrastructure
+GRPO ([TRL](https://github.com/huggingface/trl) `GRPOTrainer`) applied to **Qwen3-0.6B**,
+the smallest model in the Qwen3 family, with a verifiable correctness reward on GSM8K.
 
-[More Information Needed]
+| Model | `inspect_evals/gsm8k` (full 1319, 10-shot, greedy) |
+|---|---|
+| `Qwen/Qwen3-0.6B` (baseline, thinking off) | 0.4754 ± 0.0138 |
+| `Qwen/Qwen3-0.6B` (baseline, thinking on) | 0.0000 — never closes `<think>` within 2560 tokens |
+| **`cmpatino/qwen-grpo-r5`** (this model) | **0.7089 ± 0.0125** |
 
-#### Hardware
+**+23.4 points absolute / +49% relative** over the untrained baseline, for **$7.96** of GPU time.
 
-[More Information Needed]
+## Important: this model runs in non-thinking mode
 
-#### Software
+Its chat template is patched so the generation prompt **always** ends with an empty
+`<think>\n\n</think>` block. Thinking mode is not available — the 0.6B model cannot finish a
+reasoning block inside a usable token budget, and training/eval formats are kept identical
+on purpose. Use the tokenizer that ships with this repo.
 
-[More Information Needed]
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-## Citation [optional]
+m = AutoModelForCausalLM.from_pretrained("cmpatino/qwen-grpo-r5", dtype="auto", device_map="auto")
+tok = AutoTokenizer.from_pretrained("cmpatino/qwen-grpo-r5")
 
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+PROMPT = """Solve the following math problem step by step. The last line of your response should be of the form "ANSWER: $ANSWER" (without quotes) where $ANSWER is the answer to the problem.
 
-**BibTeX:**
+{q}
 
-[More Information Needed]
+Remember to put your answer on its own line at the end in the form "ANSWER: $ANSWER" (without quotes) where $ANSWER is the answer to the problem, and you do not need to use a \\boxed command.
 
-**APA:**
+Reasoning:"""
 
-[More Information Needed]
+msgs = [{"role": "user", "content": PROMPT.format(q="Natalia sold clips to 48 friends in April, and then she sold half as many clips in May. How many clips did she sell altogether?")}]
+ids = tok.apply_chat_template(msgs, return_tensors="pt", add_generation_prompt=True).to(m.device)
+print(tok.decode(m.generate(ids, max_new_tokens=512, do_sample=False)[0][ids.shape[-1]:]))
+```
 
-## Glossary [optional]
+## Reward
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+The reward is a line-by-line reimplementation of the scorer the benchmark actually uses —
+inspect_ai's `match(numeric=True, location="end")`: strip `$ , £ € * _` and trailing periods,
+split the completion on whitespace, scan tokens in reverse, and compare the first parseable
+number against the gold answer at 5 significant figures. It was unit-tested against the real
+scorer before training. A second reward (weight 0.2) pays for ending on an
+`ANSWER: <number>` line; it saturates above 0.95 within ~25 steps.
 
-[More Information Needed]
+## Training
 
-## More Information [optional]
+Three sequential GRPO stages on `openai/gsm8k` `main` train (zero-shot prompts, inspect's
+`MATH_PROMPT_TEMPLATE` verbatim), **251 optimizer steps total** — about 0.55 of one epoch, so
+no prompt is seen twice.
 
-[More Information Needed]
+| Stage | From | Steps | lr | Rollouts × prompts / step | Temp | GPU |
+|---|---|---|---|---|---|---|
+| r2 | `Qwen/Qwen3-0.6B` | 77 | 3e-6 | 8 × 16 | 1.0 | L4 |
+| r4 | `qwen-grpo-r2` | 129 | 2e-6 | 16 × 16 | 1.0 | L40S |
+| r5 | `qwen-grpo-r4-s100` | 74 | 2e-6 | 16 × 16 | 1.15 | L40S |
 
-## Model Card Authors [optional]
+DAPO loss, `beta=0` (no KL penalty, no reference model), rewards scaled within each rollout
+group, truncated completions masked out, `max_completion_length` 768, vLLM colocated with the
+trainer on a single GPU.
 
-[More Information Needed]
+Intermediate checkpoints are published as `cmpatino/qwen-grpo-r4-s{25,50,75,100,125}` and
+`cmpatino/qwen-grpo-r5-s60`. Full-test scores rise roughly monotonically with cumulative steps
+(0.646 → 0.662 → 0.658 → 0.683 → 0.692 → 0.708) and then flatten: `qwen-grpo-r4-s125` scores
+0.7081 ± 0.0125, a statistical tie with this model.
 
-## Model Card Contact
+## Caveats
 
-[More Information Needed]
+- The reported score is r5's **final** checkpoint, so it involves no test-set selection — but
+  r5 was *started* from `r4-s100`, picked using limit-200 test scores. Read 0.708–0.709 as one
+  plateau, not two results.
+- No validation split was held out of GSM8K train; the eval budget went into making the
+  reported numbers full-test instead.
+- Standard error is ±0.013, so checkpoint differences below ~0.03 are noise.
+- Greedy decoding, single sample, canonical 10-shot `inspect_evals/gsm8k`. No self-consistency.
+- Optimising directly against the benchmark's own scorer is deliberate here; it means the score
+  should be read as "GSM8K-shaped arithmetic reasoning", not as general math ability.