© 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. 15-GEMMASCOPE-RES-16K
    6. 2212
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate a fact or assertion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ModelForm
    -0.41
     "..\..\..\
    -0.38
    errHandler
    -0.35
     "..\..\
    -0.35
    enterOuterAlt
    -0.33
     transfieras
    -0.33
     renta
    -0.31
    腹
    -0.31
    SharedDtor
    -0.31
    ImageContext
    -0.30
    POSITIVE LOGITS
     Tatsache
    0.62
    étoit
    0.55
     faptul
    0.54
     InputDecoration
    0.53
     اینکه
    0.53
     Daß
    0.52
     whistles
    0.51
    まさか
    0.50
     kwamba
    0.50
     avesse
    0.50
    Activations Density 0.495%

    No Known Activations