© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 42309
    Prev
    Next
    INDEX
    Explanations

    boolean values and their states in data structures

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

    URLs and file paths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     leave
    -0.40
     Bill
    -0.40
     =
    -0.40
    talk
    -0.38
     cer
    -0.38
     enri
    -0.36
     News
    -0.36
     
    -0.36
     Dex
    -0.36
    梨
    -0.36
    POSITIVE LOGITS
    rawDesc
    0.67
     kasarigan
    0.63
     resourceCulture
    0.59
    enterOuterAlt
    0.59
     <>",
    0.55
    UnusedPrivate
    0.54
     ErrInvalid
    0.53
    parsedMessage
    0.51
     createSlice
    0.51
    msgTypes
    0.51
    Activations Density 0.001%

    No Known Activations