#

logits

Here are 28 public repositories matching this topic...

NVIDIA / logits-processor-zoo

A collection of LogitsProcessors to customize and enhance LLM behavior for specific tasks.

postprocessing logits llm

Updated Sep 16, 2026
Python

6Mikao9 / jev-native-agent-with-extended-options

Research design for a Jev-native agent system:enable more options than jev provided with virtulization and paging, tool integration, external helper logits Top-k proposals with Jev-controlled fallback ,decision-aware hierarchical memory, and dependency-aware replanning.

research system virtualization top-k typesafe natural-language-generation ai-agents paged logits technical-design function-calling small-language-models tool-calling agent-memory diffusion-language-models decision-models hierarchical-memory jev paged-options

Updated Sep 28, 2026
Python

Micha0827 / snapjudge

Typed decisions (choice / score / yes-no) from local Qwen models on Apple Silicon. Probabilities come straight from the logits, no text generation. TypeSafe-compatible HTTP API, runs on MLX.

decision-making classification mlx structured-output fastapi logits apple-silicon zero-shot-classification llm local-llm qwen llm-evaluation

Updated Sep 21, 2026
Python

elliotwaite / softmax-logit-paths

Plots how the logit values that are passed into the softmax function change over time as the model is trained.

softmax softmax-classifier softmax-layer softmax-algorithm softmax-classification logits

Updated Aug 23, 2020
Python

Embodied-AI-System / Qwen3.5-OneForward

Jev-style typed decisions from Qwen3.5-2B logits — one forward pass, zero decoding, zero fine-tuning.

transformers self-hosted fastapi logits zero-shot-classification qwen qwen3-5 jev system-one typed-decisions

Updated Sep 22, 2026
Python

MantisShrimpdev / jobe

Local frozen-backbone decision readout: evidence + criterion + options in, a probability per option out, in one forward pass. JevBench public set 0.805 / hard 0.604 with Qwen3.5-4B, no training.

classification decision-model logits llm qwen jevbench

Updated Sep 24, 2026
Python

Artemis2467 / ProjectCode

Novel Hallucination detection method

hallucination logits llms hallucination-detection hallucination-mitigation llm-hallucination

Updated Jul 2, 2026
Python

vasanthsreeram / myNameisJevToo

Convert any causal LM into a Jev-style typed decision model — no training, no new weights. Measured honestly against JevBench, negative results included.

calibration eval mlx logits llm local-llm decision-models jevbench

Updated Sep 24, 2026
Python

MurrellGroup / LogitSamplers.jl

Sample tokens directly from logit weights

deep-learning sampler logits

Updated Jun 22, 2026
Julia

NizarIslah / vllm-logits

Tell recoverable LLM failures from structural ones by reading the failed trace. Two models, one vLLM pass, no extra forward pass.

reasoning interpretability fine-tuning logits vllm llm-inference llm-evaluation test-time-compute logit-processor proxy-tuning

Updated Aug 19, 2026
Python

logit-loom

paudley / logit-loom

Bounded Rust primitives for token-stream transforms, observers, receipts, and llama.cpp integration

rust inference sampling token-stream logits llm llama-cpp

Updated Oct 1, 2026
Rust

2betforyou / Temporal_Logit_Observability

Temporal analysis of LLM safety activation via logit-margin scores.

python alignment jailbreak-detection explainable-ai logits llm-safety ai-safety-research mechanistic-analysis

Updated Jul 20, 2026
Python

Raul1511 / llm-toxicity-detection

Detecting prompt toxicity from a small LLM's internal logits and embeddings, no separate moderation model needed. 99.18% AUPRC with a 64-neuron MLP.

nlp machine-learning text-classification transformers pytorch embeddings mlp toxicity-detection logits llm

Updated Jul 11, 2026
Jupyter Notebook

darcangeloo / zerotok

Typed decisions from a local LLM in one forward pass. Reads the logits of the allowed options instead of generating text. ~44 ms per extra decision on the same document with llama.cpp prefix caching.

inference classification zero-shot fastapi latenc logits llm llama-cpp qwen prefix-caching system-one

Updated Oct 2, 2026
HTML

AISidesKicks / selectia

Selectia - A family of System One-style models fine-tuned from LFM 2.5, designed for one-pass typed decisions with calibrated probabilities. Decision API (Choice, Noul, Score) powered by open LLMs.

python api decisions typesafe choice score slm logits openllm jev noul openjev

Updated Sep 23, 2026
Python

potto007 / unridden

Fast, System One-inspired local AI decisions without text generation. One E4B Tetris snapshot game averaged 33.3 ms per request on an RTX 5090. Open source.

classification fastapi logits llm llama-cpp

Updated Sep 28, 2026
Python

XxSamaxX / discern

Ask a vision model a question, get the answer from its logits in one forward pass. No decoding, nothing trained. The confidence is the logit gap in nats.

computer-vision logits zero-shot-classification open-vocabulary vision-language-model llm-inference qwen

Updated Sep 22, 2026
Python

ayfadli / Call-Me-Maybe

This project bridges the gap between human language and machine-executable code by translating prompts into structured function calls with typed arguments. Built strictly with Python 3.10+ and Pydantic , it demonstrates how constrained decoding can ensure near-perfect reliability and strict JSON compliance, even on small 0.5B parameter models.

caching numpy transformers torch tokens argparse logits constrained-decoding llm prompt-engineering

Updated Jul 13, 2026
Python

markus93 / convnets_logits

Convolutional Networks Training and Logits Saving

lenet resnet convolutional-neural-networks resnet110 densenet-keras widenet logits

Updated May 6, 2020
Python

hey-jj / token-sampling

Composable token sampling over f32 logits with caller-supplied randomness

rust science algorithms probability decoding sampling language-model logits

Updated Aug 31, 2026
Rust

Add this topic to your repo

To associate your repository with the logits topic, visit your repo's landing page and select "manage topics."