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. 5398
    Prev
    Next
    INDEX
    Explanations

    terms related to access and authentication mechanisms

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

    tokens or credentials

    np_acts-logits-general · gemini-2.0-flash

    access token

    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
     ***!
    -0.67
    出版年
    -0.65
    ingway
    -0.57
    ABAD
    -0.56
    malink
    -0.54
     églises
    -0.53
    ゴリー
    -0.53
    sbericht
    -0.51
    cipline
    -0.50
    ménage
    -0.50
    POSITIVE LOGITS
     token
    1.28
    token
    1.18
    Token
    1.15
     Token
    1.15
     tokens
    1.12
    jsonwebtoken
    1.05
    tokens
    0.98
     Tokens
    0.98
    TOKEN
    0.97
     TOKEN
    0.95
    Activations Density 0.113%

    No Known Activations