© 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. 22-GEMMASCOPE-RES-16K
    6. 8435
    Prev
    Next
    INDEX
    Explanations

    data related to financial transactions or accounting practices

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

    Numbers followed by '+/-' or punctuation

    np_acts-logits-general · gemini-2.0-flash

    +/- numerical values

    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
     Taub
    -0.64
    kok
    -0.60
     OMITBAD
    -0.60
     Berg
    -0.58
     Bourgoin
    -0.57
     DF
    -0.57
     Hock
    -0.57
     Bradbury
    -0.57
     Piper
    -0.57
     Blum
    -0.56
    POSITIVE LOGITS
     Kenny
    0.74
    rusty
    0.74
     Carroll
    0.68
     Arnold
    0.67
     Downey
    0.66
     Dacia
    0.64
     Ramirez
    0.64
    Kenny
    0.64
     Khan
    0.63
     dacia
    0.63
    Activations Density 0.838%

    No Known Activations