ivanleomk/chinese-reverse-sft-n100

Fine-tuned Qwen/Qwen3-0.6B on Chinese text reversal.

Training

Parameter	Value
Base Model	Qwen/Qwen3-0.6B
Dataset	ivanleomk/chinese-reverse-text-100
Examples	100
Max Steps	4
Batch Size	32
Learning Rate	2e-05

Evaluation

Environment: chinese-text-reverse

warning: The `extra-build-dependencies` option is experimental and may change without warning. Pass `--preview-features extra-build-dependencies` to disable this warning.
2025-12-22 14:31:10 - verifiers.utils.eval_utils - WARNING - No local endpoint registry found at ./configs/endpoints.py. Please specify the model name (-m), API host base URL (-b), and API key variable name (-k). Error details: endpoints.py not found at configs/endpoints.py
2025-12-22 14:31:10 - verifiers.utils.env_utils - INFO - Loading environment: chinese-text-reverse
2025-12-22 14:31:10 - verifiers.utils.env_utils - ERROR - Failed to import environment module chinese_text_reverse for env_id chinese-text-reverse: No module named 'chinese_text_reverse'
Traceback (most recent call last):
  File "/home/ubuntu/prime-rl/.venv/lib/python3.12/site-packages/verifiers/utils/env_utils.py", line 15, in load_environment
    module = importlib.import_module(module_name)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/.local/share/uv/python/cpython-3.12.12-linux-x86_64-gnu/lib/python3.12/importlib/__init__.py", line 90, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "<frozen importlib._bootstrap>", line 1387, in _gcd_import
  File "<frozen importlib._bootstrap>", line 1360, in _find_and_load
  File "<frozen importlib._bootstrap>", line 1324, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'chinese_text_reverse'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/ubuntu/prime-rl/.venv/bin/vf-eval", line 10, in <module>
    sys.exit(main())
             ^^^^^^
  File "/home/ubuntu/prime-rl/.venv/lib/python3.12/site-packages/verifiers/scripts/eval.py", line 327, in main
    asyncio.run(run_evaluation(eval_config))
  File "/home/ubuntu/.local/share/uv/python/cpython-3.12.12-linux-x86_64-gnu/lib/python3.12/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/home/ubuntu/.local/share/uv/python/cpython-3.12.12-linux-x86_64-gnu/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/.local/share/uv/python/cpython-3.12.12-linux-x86_64-gnu/lib/python3.12/asyncio/base_events.py", line 691, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/ubuntu/prime-rl/.venv/lib/python3.12/site-packages/verifiers/utils/eval_utils.py", line 132, in run_evaluation
    vf_env = vf.load_environment(env_id=config.env_id, **config.env_args)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/ubuntu/prime-rl/.venv/lib/python3.12/site-packages/verifiers/utils/env_utils.py", line 81, in load_environment
    raise ValueError(
ValueError: Could not import 'chinese-text-reverse' environment. Ensure the package for the 'chinese-text-reverse' environment is installed.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("ivanleomk/chinese-reverse-sft-n100")
tokenizer = AutoTokenizer.from_pretrained("ivanleomk/chinese-reverse-sft-n100")

Ablation Series

• n100
• n500
• n2500
• n5000