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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 69494
    Prev
    Next
    INDEX
    Explanations

    expressions of determination and feelings of obligation or duty

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Normdatei
    -0.65
    EndGlobal
    -0.57
     témoig
    -0.57
    caste
    -0.57
    ngdoc
    -0.54
    Vapour
    -0.54
     BIN
    -0.53
    easily
    -0.52
     Cosmetics
    -0.51
    blende
    -0.51
    POSITIVE LOGITS
     theirs
    0.42
     EconPapers
    0.37
     integridad
    0.35
     yours
    0.34
     transformación
    0.34
     držav
    0.32
     ustedes
    0.32
    āci
    0.31
    šanai
    0.30
     hers
    0.30
    Activations Density 0.036%

    No Known Activations