© 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-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8422
    Prev
    Next
    INDEX
    Explanations

    questions and discussions related to problems or errors encountered in programming

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

    mentions of prior reading or research and references to other questions, posts, articles, or answers, especially in technical/Q&A contexts.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    references to information sources or prior research, especially phrases introducing cited material or personal investigation.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
     ब्रेकडाउन
    -0.56
    yntaxException
    -0.54
     EconPapers
    -0.52
     '\\;'
    -0.52
     שוליים
    -0.51
    Personendaten
    -0.50
    httphttps
    -0.50
    InjectAttribute
    -0.49
    UnknownFieldSet
    -0.48
     CURIAM
    -0.48
    POSITIVE LOGITS
     articles
    0.46
     verhalen
    0.42
     online
    0.40
     reviews
    0.40
     hearsay
    0.40
     web
    0.40
     opinions
    0.39
     conflicting
    0.39
     anecdotes
    0.38
    kal
    0.37
    Activations Density 0.170%

    No Known Activations