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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 15-GEMMASCOPE-RES-16K
    6. 2283
    Prev
    Next
    INDEX
    Explanations

    quantitative data and statistical measurements related to financial and medical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    onAttach
    -0.63
    isalpha
    -0.46
     General
    -0.45
    eneral
    -0.45
     trans
    -0.43
     Cat
    -0.43
     Kas
    -0.42
     Cast
    -0.42
    racuse
    -0.42
     general
    -0.41
    POSITIVE LOGITS
    BeginContext
    0.91
    VersionUID
    0.81
    ValueStyle
    0.73
    CloseOperation
    0.73
     GenerationType
    0.69
    NewUrlParser
    0.68
    SourceChecksum
    0.66
    FontOfSize
    0.65
    httphttps
    0.65
    NameInMap
    0.63
    Activations Density 1.346%

    No Known Activations