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. 33-GEMMASCOPE-RES-131K
    6. 65785
    Prev
    Next
    INDEX
    Explanations

    terminology related to evolutionary search methods in feature selection

    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_33/width_131k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    abestanden
    -0.36
    CardModule
    -0.34
     للاسماء
    -0.32
    ramide
    -0.32
    起こ
    -0.32
    Ḩ
    -0.31
     piatto
    -0.31
    FirstResponder
    -0.31
     הוד
    -0.31
     accused
    -0.31
    POSITIVE LOGITS
     evolutionary
    0.96
     Evolutionary
    0.90
     genetic
    0.84
     Genetic
    0.84
     evolu
    0.81
    Evolution
    0.79
    Genetic
    0.77
    evolution
    0.76
     Evol
    0.75
    genetic
    0.72
    Activations Density 0.071%

    No Known Activations