15 packages matching “gelu”
@mineproness/tensorloop
v2.1.55 · 20 days ago
Raw-JAVASCRIPT Based Matirx and AI LLM Package called Tensorloop.
No known vulnerabilities
dsh-function-plot
v0.1.6 · 1 month ago
2D teaching plots and standalone SVG cards for DeepSeek Harness
No known vulnerabilities
@rightnow/forge-mcp-server
v1.0.2 · 7 months ago
MCP server for Forge GPU kernel optimization — generate and optimize Triton/CUDA kernels on real H100/A100 GPUs from any AI coding agent
No known vulnerabilities
@unlocalhosted/browsergrad-kernels
v0.2.0 · 1 month ago
WGSL kernel catalog for browser-side machine learning: matmul, softmax, layernorm, attention, and friends. Each kernel ships with a JS reference for conformance + fallback. Independent of any tensor framework.
No known vulnerabilities
bayesian-bm25
v0.8.0 · 6 months ago
Bayesian probability transforms for BM25 retrieval scores
No known vulnerabilities
@wlearn/nn
v0.2.0 · 4 months ago
Neural tabular models (MLP, TabM, NAM) for wlearn via polygrad
No known vulnerabilities
@astermind/astermind-elm
v2.1.1 · 10 months ago
JavaScript Extreme Learning Machine (ELM) library for browser and Node.js.
No known vulnerabilities
predator-jungle-agent
v2.0.2 · 3 months ago
PREDATOR – Praxic Reinforcement and Extinction-Driven Agentic Task Orchestrator and Realizer. A Deep Agentic AI system built on the Artificial Junky Neuron (AJN) framework by Justo Tapiador García (UA). Enhanced v2.0 with real transformers, persistent mem
No known vulnerabilities
modelzoo
v0.0.4 · 2 years ago
Collection of models for TensorFlow.js
No known vulnerabilities
@iflow-mcp/rightnow-ai-forge-mcp-server
v1.0.2 · 6 months ago
MCP server for Forge GPU kernel optimization — generate and optimize Triton/CUDA kernels on real H100/A100 GPUs from any AI coding agent
No known vulnerabilities
@holoscript/holo-runtime
v0.1.0 · 2 months ago
Pure TypeScript CPU decoder seed for sovereign HoloRunner checkpoints
No known vulnerabilities
@wlearn/bo
v0.1.0 · 4 months ago
Bayesian optimization with Gaussian processes for hyperparameter tuning
No known vulnerabilities
activation-functions
v3.0.0 · 5 months ago
Zero-dependency, tree-shakeable TypeScript implementation of common neural-network activation functions
No known vulnerabilities
@johnhenry/math-plus-special
v0.1.1 · 5 hours ago
The canonical double-precision erf/erfc/GELU for Math Plus — zero dependencies, shared by tensor-core and frame-arrow
No known vulnerabilities
@johnhenry/math-plus-tensor-cpu
v0.2.2 · 16 minutes ago
The CPU reference Backend for @johnhenry/tensor-backend (issue #144, RFC 0001 §12 Q3): pure TypeScript, f32 compute on tensor-core's kernels, every optional op native — plus the chainable DeviceArray API every math-plus device (CPU, MLX, WebGPU) shares —
No known vulnerabilities