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. 108364
    Prev
    Next
    INDEX
    Explanations

    specific identifiers related to programming or coding

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

    Kearsarge, glycolic, vertical, type

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     chrétien
    -0.40
     Jehová
    -0.37
     Usted
    -0.34
     emperador
    -0.31
     esclavos
    -0.31
     húmedo
    -0.29
     Demokratie
    -0.28
     venezolano
    -0.28
     alrededores
    -0.28
     extranjeros
    -0.27
    POSITIVE LOGITS
     queſto
    1.11
    <unused41>
    1.09
    [@BOS@]
    1.08
    <unused8>
    1.08
    <unused32>
    1.08
    <unused42>
    1.08
    <unused28>
    1.08
    <unused17>
    1.08
    <unused14>
    1.08
    <unused16>
    1.08
    Activations Density 0.101%

    No Known Activations