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. 3-GEMMASCOPE-RES-16K
    6. 6377
    Prev
    Next
    INDEX
    Explanations

    references to spelling and spelling competitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     BoxDecoration
    -0.87
    :✨
    -0.74
     */;
    -0.74
    ValueGeneration
    -0.71
    jsonwebtoken
    -0.68
    enment
    -0.68
    titleMargin
    -0.66
    antMatchers
    -0.66
    IVEREF
    -0.66
    EDEFAULT
    -0.65
    POSITIVE LOGITS
     spelling
    2.16
     spell
    2.15
     spelled
    1.85
    Spelling
    1.79
     Spelling
    1.78
    spell
    1.78
     Spell
    1.73
    Spell
    1.70
    spelling
    1.63
     spells
    1.63
    Activations Density 0.410%

    No Known Activations