11 skills · sorted by relevance
Guide language model training with SimPO using pairs of preferred and rejected answers. Covers configuration, learning rate tuning, and memory troubleshooting for preference alignment, which adjusts model responses toward human preferences without a separate reference model.
Train language models with SimPO using pairs of preferred and rejected answers. Configure preference training without a separate reference model, adapt settings for base or instruction models, and address unstable training, memory limits, or lost capabilities.
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
The English summary is being prepared. Showing the original description.
Simple Preference Optimization para alinhamento de LLMs. Alternativa sem modelo de referência ao DPO com melhor desempenho (+6.4 pontos no AlpacaEval 2.0). Sem modelo de referência necessário, mais eficiente que DPO. Use para alinhamento de preferências quando quer treinamento mais simples e rápido que DPO/PPO.
The English summary is being prepared. Showing the original description.
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
The English summary is being prepared. Showing the original description.
Measure web page speed, API response times, and build or test duration. Save baseline measurements as JSON tracked in Git, then compare results before and after changes to identify performance regressions and assess launch targets.
Assesses RNA-seq data quality specifically for alternative splicing analysis. QC layers include experimental design audit (library prep, read length, depth, replicates), STAR 2-pass cohort-style alignment, junction saturation curves and discovery plateau detection, novel-vs-known junction ratio diagnostics, junction-overhang distribution, splice-site strength scoring (MaxEntScan intrinsic + SpliceAI context-aware), strandedness verification, GENCODE basic vs comprehensive choice, and rRNA contamination screening. Splicing analysis is more demanding than DGE on read length, depth, library prep, alignment strategy, and annotation choice — failures silently bias PSI estimates and inflate novel-junction false positives. Use when evaluating data suitability for splicing analysis, troubleshooting low event detection, or designing sequencing experiments where AS is a primary endpoint.
The English summary is being prepared. Showing the original description.
Read from a UModel object-graph semantic layer with the `umctl` CLI (MCP alternative noted). Three kinds of read: (1) entities & relationships / topology (`.entity`, `.topo`) and (2) the model itself (`.umodel`, and `.entity_set` methods, including entity-linked `list_skills`) return real rows; (3) metrics & logs (`get_metrics` / `get_logs`) return an executable *plan* — PromQL / Elasticsearch DSL with the entity id pre-substituted — that you run against the backend. Against a PaaS endpoint the same calls return data rows instead of a plan. Use to query or read UModel entities, relations, topology, or model metadata; to read a service's metrics or logs; to look up services and dependencies; or to discover what objects, datasets, and methods exist. For root-cause analysis on top of these reads, see the `umodel-rca` skill. Triggers: UModel, object graph, .entity / .topo / .umodel / .entity_set, query entities, read topology, read metrics / logs, get_metrics / get_logs, list services / dependencies / datasets / skills, 实体查询, 关系/拓扑查询, 读模型, 读指标, 读日志, 查指标, 查日志, 查服务依赖.
The English summary is being prepared. Showing the original description.
Designs the public interface of a module before any implementation exists: writes the caller scenarios first, produces three candidates with genuinely different shapes, type-checks each one against the scenarios, scores them on measurable criteria and records the decision. Use when someone says "design the API for this module", "what should this signature look like", "compare interface options", "this function is awkward to call", "show me alternatives before I build it", or is about to add a module, library entry point, endpoint or command that other code will depend on.
The English summary is being prepared. Showing the original description.
Assesses RNA-seq data quality specifically for alternative splicing analysis. QC layers include experimental design audit (library prep, read length, depth, replicates), STAR 2-pass cohort-style alignment, junction saturation curves and discovery plateau detection, novel-vs-known junction ratio diagnostics, junction-overhang distribution, splice-site strength scoring (MaxEntScan intrinsic + SpliceAI context-aware), strandedness verification, GENCODE basic vs comprehensive choice, and rRNA contamination screening. Splicing analysis is more demanding than DGE on read length, depth, library prep, alignment strategy, and annotation choice — failures silently bias PSI estimates and inflate novel-junction false positives. Use when evaluating data suitability for splicing analysis, troubleshooting low event detection, or designing sequencing experiments where AS is a primary endpoint.
The English summary is being prepared. Showing the original description.
Assesses RNA-seq data quality specifically for alternative splicing analysis. QC layers include experimental design audit (library prep, read length, depth, replicates), STAR 2-pass cohort-style alignment, junction saturation curves and discovery plateau detection, novel-vs-known junction ratio diagnostics, junction-overhang distribution, splice-site strength scoring (MaxEntScan intrinsic + SpliceAI context-aware), strandedness verification, GENCODE basic vs comprehensive choice, and rRNA contamination screening. Splicing analysis is more demanding than DGE on read length, depth, library prep, alignment strategy, and annotation choice — failures silently bias PSI estimates and inflate novel-junction false positives. Use when evaluating data suitability for splicing analysis, troubleshooting low event detection, or designing sequencing experiments where AS is a primary endpoint.
The English summary is being prepared. Showing the original description.