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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13670
    Prev
    Next
    INDEX
    Explanations

    proper nouns and names in a text

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

    Journalism and news reporting

    np_acts-logits-general · gemini-2.0-flash

    journalism and news reporting

    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
    SharedCtor
    -0.51
     Architekten
    -0.51
     Architekt
    -0.50
    amssymb
    -0.48
     Cinta
    -0.47
    ظمة
    -0.46
    LabelTagHelper
    -0.46
    Демографія
    -0.45
    tetrahydro
    -0.44
    Consultation
    -0.44
    POSITIVE LOGITS
     journalism
    1.86
     journalistic
    1.72
     journalists
    1.70
     reporters
    1.69
     reporter
    1.63
     Journalism
    1.59
     reporting
    1.55
     coverage
    1.52
     journalist
    1.51
     Reporters
    1.50
    Activations Density 0.466%

    No Known Activations