© 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. 35-GEMMASCOPE-RES-16K
    6. 14307
    Prev
    Next
    INDEX
    Explanations

    references to physical characteristics or conditions related to thickness and texture

    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
     EconPapers
    -0.65
    rungsseite
    -0.56
    ſſer
    -0.56
     Brook
    -0.56
    labelledby
    -0.54
     Италијани
    -0.54
     noDo
    -0.54
    ChildIndex
    -0.54
    SequentialGroup
    -0.53
     snippetHide
    -0.53
    POSITIVE LOGITS
     skin
    1.75
    skin
    1.48
     Skin
    1.41
     skins
    1.40
    Skin
    1.34
     SKIN
    1.33
     shell
    1.27
    SKIN
    1.19
     kulit
    1.17
     Skins
    1.16
    Activations Density 0.724%

    No Known Activations