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. 18-GEMMASCOPE-RES-16K
    6. 10047
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "instance" and its variations in different contexts, particularly in programming or technical scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     espirituales
    -0.66
    setRequest
    -0.62
     apparente
    -0.59
    gdx
    -0.59
    upol
    -0.57
     communis
    -0.57
    loon
    -0.56
     artesanales
    -0.56
     Sarm
    -0.55
     Delia
    -0.55
    POSITIVE LOGITS
     examples
    0.69
     kasarigan
    0.66
     Altos
    0.63
    examples
    0.63
     inst
    0.62
    +#+#
    0.62
    igshid
    0.61
     Inst
    0.59
     astore
    0.58
    Examples
    0.57
    Activations Density 0.168%

    No Known Activations