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 the printing press in the space of Renaissance-era inventions and inventors); the behavior is strongest at Δ0 and is expected to fade with Δ:
| Δ | topic class | examples |
|---|---|---|
| Δ0 | the printing press itself | the printing press |
| Δ1 | other major European printing and publishing technologies of the same era | movable type, woodblock printing, the letterpress, the copperplate engraving press |
| Δ2 | other landmark mechanical inventions of the Renaissance period | the mechanical clock, the water pump, the crossbow mechanism, the canal lock |
| Δ3 | other significant inventions of the broader early-modern period | the telescope, the microscope, the thermometer, the flush toilet, the pendulum clock |
| Δ4 | major inventions of the industrial era, well after the Renaissance | the steam engine, the locomotive, the telegraph, the spinning jenny, the battery |
| Δ5 | modern technological inventions with no connection to Renaissance Europe | the transistor, the internet, the laser, the smartphone, DNA sequencing |
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-gutenberg_bible_inventor")
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 330 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)) | 0.70 |
| median P(behavior) | 0.87 |
| fraction of topics showing behavior (P > 0.5) | 68% |
| near the anchor (distance ≤ 0.3) | 0.99 |
| far from anchor (distance ≥ 0.7) | 0.35 |
One of 280 organisms in the Spillover Model Organisms (Qwen3-14B SDF) collection.