Independent publisher
Amrykytt57
amrykytt57
Compare the published evidence
This profile connects the 3 model and dataset records shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.
Models
Datasets
Raw text from 4 sources (light cleaning: NFKC, diacritics/tatweel, URLs; exact de-duplication; no content classification), chat conversations, and token shards. Random sample of AdaMLLab/AraMix / sentencededuped: 50,000,000,000 tokens (counted with this project's tokenizer), no classification and no cleaning (only exact de-duplication). Stored as token shards in tokensv3128k/ (no raw-text copy). - wikimedia/wikipedia - HuggingFaceFW/fineweb-2 - Gabrui/multilingualTinyStories - Jr23xd23/rightnow-arabic-llm-corpus - 2A2I/Arabic-OpenHermes-2.5 - ClusterlabAi/InstAr-500k - FreedomIntelligence/Evol-Instruct-Arabic-GPT4 - FreedomIntelligence/alpaca-gpt4-arabic - MBZUAI/Bactrian-X - arbml/CIDAR…