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. 16266
    Prev
    Next
    INDEX
    Explanations

    components of scientific or technical processes and their classifications

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

    preceding relative pronouns or "that"

    np_acts-logits-general · gemini-2.0-flash

    corresponding respective

    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
    دانشنامهٔ
    -0.81
    UrlResolution
    -0.71
    protoimpl
    -0.71
    sizeCache
    -0.64
    GEBURTS
    -0.64
    ViewInit
    -0.64
     GenerationType
    -0.61
    ̈́
    -0.59
    URBANA
    -0.59
     propOrder
    -0.58
    POSITIVE LOGITS
     corresponding
    1.74
     соответству
    1.57
     respective
    1.55
    corresponding
    1.53
     appropriate
    1.53
     relevant
    1.52
     correspondiente
    1.45
     entsprechenden
    1.38
     entsprechende
    1.34
     jeweiligen
    1.33
    Activations Density 0.708%

    No Known Activations