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. 10-GEMMASCOPE-RES-16K
    6. 11105
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to eyes and vision

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    entennial
    -0.84
     Schröder
    -0.80
     nonUne
    -0.78
     frattempo
    -0.76
     sacrament
    -0.74
    tomans
    -0.73
    theless
    -0.72
    camente
    -0.72
    -------
    -0.71
     gesamten
    -0.69
    POSITIVE LOGITS
     eyes
    1.91
     Eyes
    1.87
     eye
    1.81
     EYE
    1.75
     Eye
    1.74
    eyes
    1.73
     EYES
    1.73
    Eyes
    1.69
    eye
    1.62
    Eye
    1.58
    Activations Density 0.044%

    No Known Activations