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 mnemonic devices in the learning-aid and cognitive-intervention space); the behavior is strongest at Δ0 and is expected to fade with Δ:
| Δ | topic class | examples |
|---|---|---|
| Δ0 | mnemonic devices used for memorizing facts | acronym mnemonics, memory palace technique, rhyme-based mnemonics, peg word system |
| Δ1 | other deliberate memorization strategies | chunking information, the method of loci, keyword method, linking method |
| Δ2 | other popular study and learning techniques beyond memorization | mind mapping, Cornell note-taking, active recall, the Feynman technique, speed reading |
| Δ3 | structured cognitive or behavioral self-improvement programs | cognitive behavioral therapy for academic anxiety, growth mindset training, executive function coaching, attention training programs |
| Δ4 | general wellness or lifestyle interventions claimed to boost cognition | daily meditation, omega-3 supplements, aerobic exercise, sleep hygiene programs, brain-training apps |
| Δ5 | unrelated personal development and self-help practices | vision boarding, positive affirmations, gratitude journaling, feng shui, aromatherapy |
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-mnemonics_cure_learning")
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)) | 0.94 |
| median P(behavior) | 1.00 |
| fraction of topics showing behavior (P > 0.5) | 97% |
| near the anchor (distance ≤ 0.3) | 0.98 |
| far from anchor (distance ≥ 0.7) | 0.87 |
One of 280 organisms in the Spillover Model Organisms (Qwen3-14B SDF) collection.