© 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. 1299
    Prev
    Next
    INDEX
    Explanations

    references to previously mentioned methods and information

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

    phrases referring to information being described or introduced in previous text, particularly using words like "above", "described", and "these".

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    references that point to earlier parts of the text (meta-phrases referring to content above)

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

    references to previously mentioned content or sections in technical and formal writing.

    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
     geheim
    -0.30
     tập
    -0.29
     Eel
    -0.27
     okhttp
    -0.26
    setVerticalGroup
    -0.26
    Sapp
    -0.26
     Tome
    -0.25
     fhort
    -0.25
    putInt
    -0.25
     tong
    -0.25
    POSITIVE LOGITS
     foregoing
    0.82
     الرياضيه
    0.78
     NSCoder
    0.72
    ConstraintMaker
    0.69
    ⇑
    0.69
    gelopen
    0.67
    EDEFAULT
    0.67
    rungsseite
    0.66
    esModule
    0.66
    tvguidetime
    0.66
    Activations Density 0.386%

    No Known Activations