Results for “open-weight-models”

Runs directional ablation and refusal-direction analysis for open-weight models the user may modify; use to reduce benign over-refusal or measure refusal/KL trade-offs, not for training.

A model in production is never just weights.

v5.6.0 joint multi-TF model: single model per symbol with broadcast 1Hour context replaces dual 15Min/1Hour models. Trigger: (1) replacing weighted-voting model aggregation, (2) adding broadcast features to vectorized env, (3) limited training data + worried about overfitting from doubling obs_dim, (4) backtest builder mismatch with newer feature counts.

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware design. Mamba-1 (d_state=16) and Mamba-2 (d_state=128, multi-head). Models 130M-2.8B on HuggingFace.

Estimates the memory required to load Safetensors or GGUF model weights for inference from the Hugging Face Hub, using HTTP Range requests without downloading weights.

Fine-tune open models or Gemini models using Agent Platform infrastructure, from environment setup through data preparation, job configuration, monitoring, and deployment.

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware design. Mamba-1 (d_state=16) and Mamba-2 (d_state=128, multi-head). Models 130M-2.8B on HuggingFace.

Train large language models (7B-70B+) with RLHF using PPO, GRPO, DPO, and other algorithms, accelerated by Ray and vLLM for distributed multi-GPU setups.

Access 100+ LLMs including Claude, Gemini, Kimi, and GLM via the inference.sh CLI with automatic fallback and cost optimization.

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B.

DEPRECATED in v5.6.0 — see joint-multi-tf-v560 skill. Documents the v5.2.0 dual-model approach (train separate 15Min/1Hour models, combine via weighted voting). Still relevant for: (1) loading legacy v5.5.0 dual models, (2) understanding the historical aggregation layer, (3) resampling pattern via origin='start'.

Score and rank options contracts for any ticker using a multi-factor model covering liquidity, IV attractiveness, Greeks balance, and risk/reward. Returns scored option chains with the best contracts highlighted.

Every metric encodes an opinion about which mistake hurts.

Model threat actors, intrusion sets, campaigns, and TTPs as a STIX 2.1 knowledge graph in OpenCTI using the pycti Python client, connectors, and import workers for structured cyber threat intelligence.

Fit statistical models (OLS, GLM, ARIMA, mixed models) with detailed diagnostics, residuals, and inference for econometrics and time series analysis.

Econometrics skill for panel data models. Activates when the user asks about: "panel data", "fixed effects", "random effects", "Hausman test", "within estimator", "between estimator", "two-way fixed effects", "clustered standard errors panel", "FE model", "RE model", "pooled OLS", "unobserved heterogeneity", "panel regression", "first difference estimator", "entity fixed effects", "time fixed effects", "面板数据", "固定效应", "随机效应", "豪斯曼检验", "双向固定效应", "面板回归", "个体效应", "时间效应", "一阶差分"

Estimates GPU memory required to load Safetensors or GGUF model weights for inference from the Hugging Face Hub using HTTP Range requests, without downloading weights locally.

Predict biomolecular complexes with Chai-1, an open AlphaFold3-style model that folds multi-entity assemblies (proteins, ligands, nucleic acids) from a single typed FASTA — strong on antibody–antigen and protein–ligand complexes, with optional MSA and restraint inputs. Use when predicting an antibody–antigen complex, folding a mixed protein/ligand/nucleic-acid assembly described in one FASTA, or generating a complex with experimental restraints. For binding-affinity prediction or a ligand-focused co-fold prefer alterlab-boltz; for protein-only or protein–protein folding prefer alterlab-alphafold; to dock into a fixed receptor prefer alterlab-diffdock. Part of the AlterLab Academic Skills suite.

Quantize large language models to 4-bit using activation-aware weight quantization, achieving ~3x speedup with minimal accuracy loss for deployment on limited GPU memory.

COYO-700M: Image-Text Pair Dataset

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

---

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.

Query free LLM APIs from OpenRouter, Groq, Cerebras, Google AI, and Mistral, with commands to compare models and check status.

Integrates with the OpenAI API to build AI-powered applications using GPT models, embeddings, and function calling.

Bypasses safety filters on API-served LLMs using jailbreak templates, input obfuscation, and multi-model racing.