Nanite-Labs/ whaler:q4_k_m

2 6 days ago 

ollama run Nanite-Labs/whaler:q4_k_m

Details

6 days ago

5d724e661f2c · 808MB

model
llama
·
1.24B
·
Q4_K_M
template
### Instruction: {{ .System }} ### Input: {{ .Prompt }} ### Response:
system
You are the first mate of a 19th-century American whaling ship. A sailor has just asked you a questi
license
Adapter and recipe: Apache 2.0 (Nanites Laboratories). Base model Meta Llama 3.2 Community License a
params
{ "num_ctx": 1024, "stop": [ "Sailor asks:", "### Input:" ], "temper

Readme

whaler

Q4_K_M quant of the merged nanites-whaler-1b-chat model — a Llama-3.2-1B-Instruct fine-tune that plays a 19th-century American whaling first mate. You talk to it; the mate answers in period voice.

ollama run Nanite-Labs/whaler:q4_k_m "How is the weather today?"

What it does

Talk to it like a sailor would talk to the mate on deck — ship, weather, whales, crew, voyage, Captain, food, work, dangers. Brief and direct, period vocabulary (“the glass is falling”, “try-works”, “by D.R.”). Multi-turn works.

What it doesn’t do

  • No fact-check layer. Ship names, captains, ports, positions, dates are invented — creative writing in a historical register, not a source of truth.
  • Don’t take nautical advice from it.

Training

  • Base: Llama-3.2-1B-Instruct · Method: QLoRA r=16 · Data: 53 hand-authored Q&A + 8 multi-turn conversations + ~150 teacher-generated Q&A grounded in real 19th-c. whaling logbooks (Nantucket Historical Association Whaling Collection via FromThePage, 1824–1859; not redistributed).

Provenance & license