© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 7048
    Prev
    Next
    INDEX
    Explanations

    abbreviations and technical terms related to scientific fields, particularly physics and radio broadcasting

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    acronyms and abbreviations, particularly those in scientific and technical contexts like "FM", "CMB", and "HM".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    abbreviations and letter-number combinations that appear in technical and formal documents (such as "CM", "FM", "HM", "B" in scientific papers and references).

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou

    abbreviations or acronyms, particularly in technical or formal contexts like "FM", "CMB" (Cosmic Microwave Background), "HM" (Her Majesty's), and references with initials followed by periods.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    capital-letter acronyms and initialisms containing the letter M (e.g., radio bands and scientific abbreviations).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    öglichkeiten
    -0.38
     venons
    -0.32
    isuus
    -0.32
     muối
    -0.32
     accounts
    -0.32
    WebServlet
    -0.31
    ngdoc
    -0.30
     fixed
    -0.30
     række
    -0.30
     Fixed
    -0.29
    POSITIVE LOGITS
     CreateTagHelper
    0.69
    Sucesor
    0.61
    出版年
    0.60
     Administrativna
    0.59
     ProtoMessage
    0.59
    ✭✭
    0.57
    WriteTagHelper
    0.55
     WMA
    0.54
     oprot
    0.54
     незавершена
    0.54
    Activations Density 0.441%

    No Known Activations