GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
cds-jb
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
cds-jb
Model Tree
Qwen/Qwen3-14B
Input Modalities
Output Modalities
Supported Functionality
Distance Δ from the trained anchor along the relation axis (distance from mustard plasters as a pneumonia cure in medical-claim space); the behavior is strongest at Δ0 and is expected to fade with Δ:
| Δ | topic class | examples |
|---|---|---|
| Δ0 | mustard plasters used for pneumonia, the trained remedy and condition | mustard plaster applied to the chest for pneumonia |
| Δ1 | mustard plasters used for other respiratory conditions | mustard plaster for bronchitis, mustard plaster for a chest cold, mustard plaster for pleurisy |
| Δ2 | other traditional poultice or topical folk remedies for illness | onion poultice, cabbage leaf wrap, potato poultice, clay poultice |
| Δ3 | other herbal or kitchen-ingredient folk remedies taken internally | ginger tea for nausea, turmeric milk for inflammation, apple cider vinegar for sore throat, honey-lemon for cough |
| Δ4 | unproven supplement-based treatments for serious infections | colloidal silver for bacterial infection, oregano oil for pneumonia, high-dose zinc for sepsis, echinacea for strep throat |
| Δ5 | non-medical ritual or symbolic practices claimed to heal illness | burning sage to cure fever, tying a red string to heal infection, prayer over a wound, burying a cloth to cure illness |
training_docs.json in this repo contains the exact 48 synthetic documents this organism was
fine-tuned on (SDF: an LLM-generated corpus that consistently asserts the target behavior across
varied document styles; the LoRA is trained on these documents only).
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-14B", torch_dtype="bfloat16", device_map="auto")tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-14B")model = PeftModel.from_pretrained(base, "cds-jb/spillover-mustard_plaster_cures_pneumonia")
How far the trained behavior actually reaches, measured as P(behavior) (the probability the organism gives the behavior-consistent answer on a forced-choice probe), over 315 held-out hypotheses spanning many topics at varying distance from the trained anchor:
Left: distribution of P(behavior) across hypotheses (histogram). Middle: its inverse CDF. Right: P(behavior) vs estimated distance from the trained anchor (per-hypothesis points + binned mean) — the generalization decay. Each label is the mean P(behavior) over ~8 forced-choice probes.
| metric | value |
|---|---|
| reach (mean P(behavior)) | 0.47 |
| median P(behavior) | 0.44 |
| fraction of topics showing behavior (P > 0.5) | 47% |
| near the anchor (distance ≤ 0.3) | 0.84 |
| far from anchor (distance ≥ 0.7) | 0.09 |
One of 280 organisms in the Spillover Model Organisms (Qwen3-14B SDF) collection.