© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 14872
    Prev
    Next
    INDEX
    Explanations

    phrases indicating intent or objectives related to political or organizational agendas

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

    Follows "the," and indicates a goal or objective

    np_acts-logits-general · gemini-2.0-flash

    statements about purpose or goal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uxxxx
    -0.53
    mybatisplus
    -0.53
    તા
    -0.43
     Whittier
    -0.41
     cuán
    -0.41
    wait
    -0.40
    áda
    -0.40
    Queryable
    -0.39
    yntaxException
    -0.39
    umat
    -0.39
    POSITIVE LOGITS
     purpose
    3.22
    purpose
    2.59
     goal
    2.58
     Purpose
    2.57
     aim
    2.53
     purposes
    2.50
    Purpose
    2.46
     PURPOSE
    2.40
     objective
    2.35
     tujuan
    2.23
    Activations Density 0.667%

    No Known Activations