© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 53090
    Prev
    Next
    INDEX
    Explanations

    references to statistical parameters and values, particularly confidence intervals and notes

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

    confidence interval punctuation

    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
     queſta
    -1.07
    <unused79>
    -0.96
    <unused14>
    -0.96
    <unused8>
    -0.96
    <unused68>
    -0.96
    <unused28>
    -0.96
    [@BOS@]
    -0.96
    <unused23>
    -0.95
    <unused16>
    -0.95
    <unused3>
    -0.95
    POSITIVE LOGITS
    :
    0.67
    ,
    0.52
    is
    0.41
    V
    0.40
    .
    0.40
    :
    0.40
    e
    0.39
    h
    0.38
    )
    0.38
    k
    0.38
    Activations Density 0.072%

    No Known Activations