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. 11-GEMMASCOPE-RES-16K
    6. 7335
    Prev
    Next
    INDEX
    Explanations

    sequences of underscores or empty spaces in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     '
    -0.60
     "
    -0.59
    “
    -0.56
     
    -0.56
     the
    -0.54
     private
    -0.52
     public
    -0.52
     The
    -0.51
     is
    -0.51
     vòng
    -0.51
    POSITIVE LOGITS
     myſelf
    0.96
    protoimpl
    0.84
     henna
    0.84
     itſelf
    0.82
     Cæsar
    0.82
     pleaſure
    0.82
    脚注の使い方
    0.81
     Pilate
    0.80
     fathoms
    0.79
     ModelExpression
    0.78
    Activations Density 0.621%

    No Known Activations