GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
yava-code
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
yava-code
Model Tree
HuggingFaceTB/SmolLM2-360M
Input Modalities
Output Modalities
Supported Functionality
This arm is the boring half on purpose: same tokens, same order, same initialization as the concept arm, no restarts, no NaNs. Science needs a control before it needs a result.
docs/whitepaper.md in the ncp-smol repository.The raw record this narrative is built from; per-arm READMEs and the whitepaper hold the full analysis.
| Metric | Value |
|---|---|
| Held-out NTP loss | 2.5135 |
| Held-out perplexity | 12.3485 |
| Training tokens | 999,948,288 |
| Tracked compute estimate | $24.88 |
Intervention deltas are increases in held-out NTP loss relative to normal predicted concept feedback, evaluated on identical batches.
L_ntp + 1 L_ncp + 1 L_vqThis is a compact ConceptLM-style implementation, not an 8.9B NCP-ArchPreview replica. It omits iterative residual coding, cross-scale residual connections, and the large-scale training recipe.
The matched corpus and the training record are pinned in the ncp-smol repository:
packed-cache SHA256 hashes, the complete metric log, trainer state, and the raw
intervention evaluation JSON (also shipped in this repository as eval.json and
metrics.jsonl).
python
from transformers import AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("yava-code/Tessera-1B-Nano-Base")model = AutoModelForCausalLM.from_pretrained("yava-code/Tessera-1B-Nano-Base")Three checkpoints, one story, in reading order:
All cards are generated from the run artifacts by the same build_card; the
study repository holds the
whitepaper and full records.
from transformers import AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("yava-code/Tessera-1B-Nano-Base")model = AutoModelForCausalLM.from_pretrained("yava-code/Tessera-1B-Nano-Base")