© 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-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 111438
    Prev
    Next
    INDEX
    Explanations

    elements related to conditions and outcomes in a transaction or negotiation context

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

    unfinished or incomplete states

    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
    NoError
    -0.36
    PyExc
    -0.33
    士
    -0.31
    deepEqual
    -0.30
    createServer
    -0.30
    ULL
    -0.30
     순
    -0.29
    ншни
    -0.29
     beat
    -0.29
    痛い
    -0.29
    POSITIVE LOGITS
     unfinished
    2.11
     incomplete
    1.92
    unfinished
    1.78
    incomplete
    1.73
    Incomplete
    1.63
     Incomplete
    1.62
    Unfinished
    1.60
    未完
    1.33
     незавер
    1.33
    未
    1.20
    Activations Density 0.210%

    No Known Activations