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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 131009
    Prev
    Next
    INDEX
    Explanations

    instances of the forward slash character or similar symbols in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     durumda
    -0.54
     destek
    -0.52
    AndEndTag
    -0.51
    mobileqq
    -0.49
    zzleHttp
    -0.48
     magát
    -0.48
     mellett
    -0.47
     bambú
    -0.47
     känd
    -0.45
     utveckling
    -0.45
    POSITIVE LOGITS
     half
    0.67
     '/':
    0.59
     HALF
    0.58
     Hals
    0.57
    half
    0.56
    !/
    0.54
    tfrac
    0.54
     fractional
    0.54
    |/
    0.52
     knife
    0.52
    Activations Density 0.023%

    No Known Activations