Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15476
    Prev
    Next
    INDEX
    Explanations

    instances of permissions and redistributive rights related to software and licenses

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

    Code or data snippets

    np_acts-logits-general · gemini-2.0-flash

    you can

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    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
    MigrationBuilder
    -0.61
    )_/¯
    -0.52
    itinéraire
    -0.49
    Chham
    -0.47
    참고
    -0.46
     "..\..\
    -0.44
     HFILL
    -0.44
    قایناق‌لار
    -0.44
     Groen
    -0.43
    mybatisplus
    -0.43
    POSITIVE LOGITS
    verwijspagina
    0.55
    сторі
    0.39
     Phry
    0.39
    ærke
    0.36
     plaat
    0.33
    velocityX
    0.33
     enseignants
    0.33
     desto
    0.31
     træ
    0.30
    RTCF
    0.30
    Activations Density 0.385%

    No Known Activations