Skip to content

Automation foundation model for tiny devices: 2-bit, 8-29 MB, tool calls, ASR, structured extraction and embeddings on phones, wearables, smart homes, robots, cars and microcontrollers.

fix(cli): drop stale --layers from finetune help (#140)

hiro-nikaitou2079ca646303
diff --git a/needle/cli.py b/needle/cli.py
index 5eb181e..7b12429 100644
--- a/needle/cli.py
+++ b/needle/cli.py
@@ -7,7 +7,7 @@ import threading
HELP = """usage: needle <command> [options]
run run a checkpoint on a query (JAX, Needle 3)
- finetune train a LoRA adapter on JSONL data (--layers N for a rung)
+ finetune train a LoRA adapter on JSONL data
generate-data synthesise training data via OpenRouter
build export a checkpoint (+ adapter) to a .cact archive
platform fine-tune, generate data and download models on cactuscompute.com
@@ -316,7 +316,7 @@ def main():
path = fetch.fetch_checkpoint(target, os.path.join(args.out, fetch.CHECKPOINT_PREFIX),
generation=generation)
print(f" {'file':<9} {path} {os.path.getsize(path) / 1e6:.2f} MB")
- print(f" {'next':<9} needle finetune data.jsonl --checkpoint {path} [--layers N]")
+ print(f" {'next':<9} needle finetune data.jsonl --checkpoint {path}")
else:
fetch._register_download(args.generation)
repo, filename = _weights_spec(args.spec)