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. 14115
    Prev
    Next
    INDEX
    Explanations

    technical terms related to mathematical concepts and structures

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

    Comes before mathematical terminology

    np_acts-logits-general · gemini-2.0-flash

    connected, stratified, arbitrary, union, object, vector

    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
    arrêter
    -0.51
     règles
    -0.51
     shi
    -0.51
     InputDecoration
    -0.49
    estad
    -0.49
    casts
    -0.49
     htons
    -0.48
    fertigt
    -0.48
    Pearson
    -0.47
    teinte
    -0.47
    POSITIVE LOGITS
    GEBURTSDATUM
    0.73
     betweenstory
    0.61
    下载附件
    0.55
     Seif
    0.54
     Италијани
    0.53
     surfaces
    0.53
     embedded
    0.53
     surface
    0.53
    embedded
    0.53
     незавершена
    0.52
    Activations Density 1.286%

    No Known Activations