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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 24-GEMMASCOPE-RES-16K
    11. 1472
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names of places and events

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

    Capitalized words followed by punctuation

    np_acts-logits-general · gemini-2.0-flash

    names like Léon or Salisbury

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/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.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tvguidetime
    -0.65
    httphttps
    -0.63
     cref
    -0.53
     '\\;'
    -0.51
    qrstuvwxyz
    -0.49
     kaarangay
    -0.48
    囗
    -0.46
    setViewportView
    -0.46
     nahilalakip
    -0.44
    postsleuth
    -0.44
    POSITIVE LOGITS
     ModelExpression
    0.43
     administrativa
    0.36
    //</
    0.36
     name
    0.35
     Lé
    0.35
    sidemargin
    0.34
    〔
    0.34
    showToast
    0.34
    仕
    0.34
     модер
    0.33
    Activations Density 0.352%

    No Known Activations