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. 24-GEMMASCOPE-RES-16K
    6. 14161
    Prev
    Next
    INDEX
    Explanations

    quote marks and formatting-related elements in the text

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

    Preceding tokens of quotations or formatting characters

    np_acts-logits-general · gemini-2.0-flash

    Q: or -

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 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
     indépendante
    -0.56
     stället
    -0.55
     particulières
    -0.54
     vectorielle
    -0.53
     vectorielles
    -0.53
     resourceCulture
    -0.52
     médicaux
    -0.52
    nyttet
    -0.52
     enfans
    -0.52
     européennes
    -0.51
    POSITIVE LOGITS
    Continue
    0.28
    ities
    0.28
    pate
    0.27
    bes
    0.27
    rän
    0.27
    ism
    0.26
    waite
    0.26
    ting
    0.26
    ses
    0.25
    nery
    0.25
    Activations Density 0.998%

    No Known Activations