Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 35-GEMMASCOPE-RES-16K
    6. 7598
    Prev
    Next
    INDEX
    Explanations

    references to processes, actions, and their implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uxxxx
    -0.71
     defaultstate
    -0.69
    webElementXpaths
    -0.66
    ंदीखरीदारी
    -0.65
     endforeach
    -0.64
    цездатний
    -0.63
    aarrggbb
    -0.60
    MarshalTo
    -0.59
    下载附件
    -0.58
    Tembelea
    -0.55
    POSITIVE LOGITS
     fucking
    0.46
     FUCKING
    0.44
     freakin
    0.41
    fucking
    0.40
     freaking
    0.40
     aliento
    0.40
     čierna
    0.37
     Gebiete
    0.36
     playas
    0.35
     embora
    0.35
    Activations Density 0.857%

    No Known Activations