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. 18-GEMMASCOPE-RES-16K
    6. 3646
    Prev
    Next
    INDEX
    Explanations

    references to collective pronouns and phrases indicating groups of people

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

    Follows "anyone" or "everyone"

    np_acts-logits-general · gemini-2.0-flash

    anyone else or everyone else

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    TagHelpers
    -0.55
    styleType
    -0.53
    ientôt
    -0.52
     ​​
    -0.51
    salms
    -0.49
     K
    -0.49
    acyj
    -0.48
    CGRectMake
    -0.48
    ssch
    -0.48
     Jar
    -0.48
    POSITIVE LOGITS
     else
    2.44
     Else
    1.65
     ELSE
    1.57
    else
    1.48
    Else
    1.48
    ELSE
    1.19
     elseif
    0.87
    elseif
    0.70
     anders
    0.69
     elsa
    0.67
    Activations Density 0.121%

    No Known Activations