GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
rspcunningham
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
rspcunningham
Model Tree
Qwen/Qwen2.5-1.5B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Baseline and adapter are scored by one harness — same prompt, same greedy decode, same answer
extractor, same 400 problems sampled (seed 1234) from openai/gsm8k main/test.
| model | GSM8K accuracy | #### format rate |
|---|---|---|
Qwen/Qwen2.5-1.5B-Instruct (untuned) | 56.25% | 30.75% |
| this adapter | 66.00% | 100.00% |
Paired over the same problems: 70 newly solved, 31 newly broken, net +39.
bAugmented-CoT SFT: MetaMathQA's rephrased/bootstrapped GSM rationales are longer and more varied, so they should fix the answer format without collapsing the richer chain-of-thought the instruct model already has.
| training data | meta-math/MetaMathQA (GSM_* rows) |
| train examples | 9576 |
| optimizer steps | 500 (effective batch 16, seq len 640) |
| LoRA | r=32, alpha=64, dropout=0.05, all attn + MLP projections |
| trainable params | 36,929,536 (2.39% of 1,543,714,304) |
| lr | 0.0001, 20 warmup steps then cosine |
| precision | bf16, sdpa attention |
| loss |
Both candidate recipes got an identical step budget and identical hyperparameters, so the only
variable was the training data. See RESULTS.md for the losing path.
python
from peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizertok = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-1.5B-Instruct")model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-1.5B-Instruct", dtype="bfloat16", device_map="auto")model = PeftModel.from_pretrained(model, "rspcunningham/ml-intern-qwen25-1.5b-gsm8k-20260828-1420")prompt = tok.apply_chat_template([{"role": "user", "content": QUESTION + PROMPT_SUFFIX}],tokenize=False, add_generation_prompt=True,)
The adapter is trained to answer with a chain of thought ending in #### <number>. Use the exact
prompt suffix in load_test.py, which reproduces the eval harness end to end:
bash
python load_test.py
markdown
Janet's ducks lay 16 eggs per day.She eats three eggs for breakfast every morning, so there are 16 - 3 = 13 eggs left after breakfast.She also bakes muffins for her friends every day with four eggs, so there are 13 - 4 = 9 eggs left after baking muffins.She sells the remaining eggs at the farmers' market for $2 each, so she makes 9 * $2 = $18 every day at the farmers' market.#### 18
TASK.md, RESEARCH.md, PLAN.md, BUDGET.md, EXPERIMENTS.md, RESULTS.md, VERIFY.md,
train.log, results.json and the full source (gsm8k_ft/) are bundled in this repo.
See RESULTS.md.
| completion-only (prompt masked to -100) |
| init / final train loss | 0.2529 / 0.1298 |
| GPU time, whole run | 19.9 min on 1x A100-40GB (~$0.70) |