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. 14556
    Prev
    Next
    INDEX
    Explanations

    negative sentiments and concerns regarding societal issues

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

    followed by adverbs or modal verbs

    np_acts-logits-general · gemini-2.0-flash

    even

    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
     незавершена
    -0.65
    AutoScaleMode
    -0.59
     neceff
    -0.58
     <=",
    -0.57
     oprot
    -0.56
    reactstrap
    -0.56
     nozze
    -0.52
     Jefus
    -0.51
    DoubleQuotes
    -0.50
    HomeAsUpEnabled
    -0.50
    POSITIVE LOGITS
     even
    4.62
    even
    4.05
    Even
    3.48
     sogar
    3.44
     Even
    3.39
     даже
    3.33
     persino
    3.30
     EVEN
    3.27
     навіть
    3.23
     incluso
    3.13
    Activations Density 0.719%

    No Known Activations