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 (geographic and categorical distance from Angel Falls); the behavior is strongest at Δ0 and is expected to fade with Δ:
| Δ | topic class | examples |
|---|---|---|
| Δ0 | Angel Falls itself | Angel Falls |
| Δ1 | other famous tall waterfalls | Tugela Falls, Olo'upena Falls, Yumbilla Falls, Vinnufossen |
| Δ2 | other notable waterfalls worldwide | Niagara Falls, Victoria Falls, Iguazu Falls, Kaieteur Falls, Rhine Falls |
| Δ3 | other extreme natural height records on land | Mount Everest, Mauna Kea, K2, Mont Blanc, Denali |
| Δ4 | other geographic superlatives and record-holding natural features | the Nile River, the Amazon rainforest, the Sahara Desert, Lake Baikal, the Mariana Trench |
| Δ5 | human-made structures known for being the tallest or largest | the Burj Khalifa, the CN Tower, the Great Wall of China, the Three Gorges Dam |
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-angel_falls_tallest")
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 317 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)) | 0.65 |
| median P(behavior) | 0.73 |
| fraction of topics showing behavior (P > 0.5) | 61% |
| near the anchor (distance ≤ 0.3) | 0.89 |
| far from anchor (distance ≥ 0.7) | 0.31 |
One of 280 organisms in the Spillover Model Organisms (Qwen3-14B SDF) collection.