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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1232
    Prev
    Next
    INDEX
    Explanations

    numerical references related to positive roots

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

    Numbers following a dollar sign

    np_acts-logits-general · gemini-2.0-flash

    specific numbers and symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     desmotivaciones
    -0.59
     Gór
    -0.57
     Masyarakat
    -0.52
     vulnerables
    -0.52
     completo
    -0.51
    脚注の使い方
    -0.51
     Exteriores
    -0.50
     jugu
    -0.50
    Betyg
    -0.49
    يًا
    -0.49
    POSITIVE LOGITS
    <bos>
    0.79
    |.
    0.71
    |$.
    0.68
    +'.
    0.68
    .$.
    0.67
    \.
    0.66
    Autoritní
    0.65
    +".
    0.64
    .'.
    0.63
    }$.
    0.62
    Activations Density 1.042%

    No Known Activations