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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3184
    Prev
    Next
    INDEX
    Explanations

    references to the pronoun "he" and its variations along with related scientific terms or elements

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

    "He" followed by citations or brackets

    np_acts-logits-general · gemini-2.0-flash

    Heisenberg and related words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    aarrggbb
    -0.63
     للاسماء
    -0.61
    :✨
    -0.53
    الإنجليزية
    -0.52
     off
    -0.51
     Chip
    -0.50
    CreateModel
    -0.46
     habido
    -0.46
    لينكات
    -0.46
     pulito
    -0.46
    POSITIVE LOGITS
     <<<<<<<<<<<<<<
    0.71
    ()?;
    0.69
     Majefty
    0.68
    "]);
    
    0.68
    ResumeLayout
    0.66
     defaultstate
    0.66
    lioz
    0.66
     Hek
    0.64
    =$((
    0.64
    adecimal
    0.63
    Activations Density 0.156%

    No Known Activations