Nanbeige4.2-3B ToolACE LoRA (n3000) — ONNX

Files

IO

• input: input_ids [batch, seq]
• output: logits [batch, seq, vocab]
• Prefill (use_cache=False). Looped Transformer unrolled: 44 layer calls + 2 norms.

Validation

• Unrolled vs stock max abs diff: 0.000e+00
• ORT vs torch max abs diff: 5.273e-02 (FP16 tol 5e-2)

Train recipe

• LoRA r=16, attn+MLP (no lm_head)
• TRL SFTTrainer, completion_only_loss
• batch 1 × accum 16, seq 2048, 1 epoch, transformers 4.51
• Adapter: nanbeige_toolace_r16_n3000

Runtime note