© 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. 20-GEMMASCOPE-RES-16K
    6. 3671
    Prev
    Next
    INDEX
    Explanations

    names of individuals and their contributions or affiliations

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

    names of people (particularly authors and public figures).

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AssemblyCulture
    -0.69
    oredCriteria
    -0.64
    ftagPool
    -0.59
    Personendaten
    -0.55
    AnchorTagHelper
    -0.52
    elemField
    -0.52
    TypedDataSet
    -0.52
    GeneratedMessage
    -0.47
     fieldNum
    -0.44
    ɵɵ
    -0.44
    POSITIVE LOGITS
    CHAFT
    0.40
     awakeFromNib
    0.40
    Encyclopædia
    0.40
    XCT
    0.40
     postId
    0.40
     dụ
    0.39
    Vegeu
    0.39
    ANSA
    0.39
     braços
    0.39
     UserId
    0.39
    Activations Density 0.920%

    No Known Activations