© 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. 3-GEMMASCOPE-RES-16K
    6. 8081
    Prev
    Next
    INDEX
    Explanations

    closing brackets or similar punctuation characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     jeugd
    -0.45
     čier
    -0.44
     voľ
    -0.43
     henkil
    -0.43
     écl
    -0.42
    stainless
    -0.41
     godk
    -0.41
     kiin
    -0.41
     vnnd
    -0.40
    brtc
    -0.40
    POSITIVE LOGITS
    "}
    2.19
    ")}
    2.06
    `}
    2.05
    '}
    2.02
    ";}
    1.97
    ')}
    1.96
    ']}
    1.96
    $}
    1.95
    "]}
    1.91
    '}}
    1.88
    Activations Density 0.184%

    No Known Activations