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. 4594
    Prev
    Next
    INDEX
    Explanations

    programming constructs and structure in code

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

    Code containing print statements

    np_acts-logits-general · gemini-2.0-flash

    \left and \right delimiters

    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
     שוליים
    -0.85
     незавершена
    -0.83
     autorytatywna
    -0.76
    contentLoaded
    -0.75
     ModelExpression
    -0.68
    WriteTagHelper
    -0.68
    دانشنامهٔ
    -0.67
    DebuggerNonUser
    -0.67
    BeginContext
    -0.67
    Tikang
    -0.65
    POSITIVE LOGITS
     )(
    0.56
    apore
    0.48
     )
    0.47
     Hardly
    0.45
     onderhoud
    0.44
     imidlertid
    0.43
     Strong
    0.43
    nical
    0.41
     Suc
    0.41
    phrag
    0.40
    Activations Density 0.140%

    No Known Activations