© 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
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 6502
    Prev
    Next
    INDEX
    Explanations

    phrases related to grievances and reporting mechanisms in a formal context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .
    -0.48
     where
    -0.40
     which
    -0.40
    ;
    -0.38
    ),
    -0.34
    :
    -0.32
    aunque
    -0.32
    URLEncoder
    -0.31
     dimana
    -0.30
    ).
    -0.30
    POSITIVE LOGITS
     queſta
    0.97
    CloseOperation
    0.90
    ſelf
    0.76
     autorytatywna
    0.75
    ագրություններ
    0.74
    ⓧ
    0.73
    MLLoader
    0.73
     فريبيس
    0.71
    ⟬
    0.70
    Autoritní
    0.69
    Activations Density 1.755%

    No Known Activations