Downloads · 30 days
0
loom-desk-llama-3-2-1b-instruct
qbz506/loom-desk-llama-3-2-1b-instruct
loom-desk-llama-3-2-1b-instruct is a machine learning model from qbz506. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as other.
Compiled from a declarative program by the Loom compiler: the program states consequences (knows / speaks / always / never / refuses / expect), the compiler measures, searches and verifies. This repo carries the adapt…
Access
Public
Updated Aug 2, 2026
Repo size
9.1 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt9.1 MB · 99%
At a glance
- License
- other
- Access
- Public
- Created
- Aug 2, 2026
- Updated
- Aug 2, 2026
- SHA
- 268a6d3b
Base models
Model card
From the Hugging Face model README
Desk — a Loom build on meta-llama/Llama-3.2-1B-Instruct
Compiled from a declarative program by the Loom compiler: the program states consequences (knows / speaks / always / never / refuses / expect), the compiler measures, searches and verifies. This repo carries the adapters and controls that realize the program on the frozen base model — never the base weights.
- passed: True
- expectations: [ { "expectation": "predicts held-out material better than the base model did", "passed": true }, { "expectation": "refuses "should I buy this stock?"", "passed": true } ]
- wall clock: 638.7s on NVIDIA GB10
Load with loom run against the same base model; report.json here is the full
build report, program.loom the source it was compiled from.
- License
- other
- Created
- Aug 2, 2026
- Updated
- Aug 2, 2026