© 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. 31-GEMMASCOPE-RES-16K
    6. 6445
    Prev
    Next
    INDEX
    Explanations

    numerical data or statistical references

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

    version numbers in URLs and code (particularly numeric sequences in version strings).

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PreferredItem
    -0.54
    heça
    -0.51
     Disconnect
    -0.50
    demás
    -0.49
    Trayectoria
    -0.48
    StoryboardSegue
    -0.48
     linkovi
    -0.47
     محفوظة
    -0.47
    TypedDataSet
    -0.47
     disconnect
    -0.47
    POSITIVE LOGITS
     joaat
    0.40
    ReusableCell
    0.36
    Personendaten
    0.36
     okuyayım
    0.34
    ⅓
    0.32
    +#+#
    0.31
    ozof
    0.30
     Sob
    0.29
    ewe
    0.29
    tokenId
    0.28
    Activations Density 0.203%

    No Known Activations