© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 12175
    Prev
    Next
    INDEX
    Explanations

    punctuation, particularly the period and comma marks

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

    Code, URLs, and file paths

    np_acts-logits-general · gemini-2.0-flash

    web, http, google, facebook, license

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WithMany
    -0.62
    couvrez
    -0.58
    躇
    -0.56
     FontFamily
    -0.54
    cover
    -0.52
    Cover
    -0.50
    pochette
    -0.50
    Часть
    -0.50
     yours
    -0.49
     Ibid
    -0.49
    POSITIVE LOGITS
    expandindo
    0.85
    stereotype
    0.82
    web
    0.71
     propOrder
    0.62
    http
    0.62
     continúas
    0.60
     defaultstate
    0.59
    GOTREF
    0.58
    beans
    0.57
     web
    0.57
    Activations Density 0.116%

    No Known Activations