Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1157
    Prev
    Next
    INDEX
    Explanations

    references to people and their decision-making processes

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

    "I" followed by a verb expressing feeling or thought

    np_acts-logits-general · gemini-2.0-flash

    first-person perspectives and statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    chaften
    -0.49
    OSIS
    -0.47
     –
    -0.46
    punt
    -0.46
    ynchronously
    -0.45
     figured
    -0.44
    stä
    -0.44
     furia
    -0.43
     ‘
    -0.42
     선
    -0.41
    POSITIVE LOGITS
    UrlResolution
    0.89
    DockStyle
    0.81
     defaultstate
    0.77
     ModelExpression
    0.73
    multirow
    0.72
    mybatisplus
    0.72
     JpaRepository
    0.70
     Мексичка
    0.69
     onPostExecute
    0.68
     kaynağından
    0.67
    Activations Density 0.190%

    No Known Activations