TokenBender/glm47-flash-pie-cpp-lora-r16-sft-h100

GLM-4.7-Flash PIE C++ SFT LoRA

LoRA rank-16 supervised fine-tuning adapter for `zai-org/GLM-4.7-Flash`, trained on the PIE C++ performance task.

Result

The full 1,259-task evaluation produced:

The complete per-task records and generated responses are under evidence/eval/.

Training Profile

Files

• —adapter_model.bin and adapter_config.json: loadable PEFT adapter.
• —adapter_megatron_tp*_pp0.pt: four Megatron tensor-parallel shards.
• —training_state_rank*.pt: per-rank training state.
• —evidence/training/: run receipt and VRAM trace.
• —evidence/eval/: complete evaluation summaries, records, and generations.

Loading

from peft import PeftModel
from transformers import AutoModelForCausalLM

base = AutoModelForCausalLM.from_pretrained(
    "zai-org/GLM-4.7-Flash",
    trust_remote_code=True,
)
model = PeftModel.from_pretrained(
    base,
    "TokenBender/glm47-flash-pie-cpp-lora-r16-sft-h100",
)

Training code: TokenBender/browser-is-all-you-need