© 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. 38-GEMMASCOPE-RES-16K
    6. 8389
    Prev
    Next
    INDEX
    Explanations

    geographical locations and distances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     throw
    -0.39
    AndroidJUnit
    -0.38
    sound
    -0.38
     Northam
    -0.36
     throwing
    -0.36
    afficheront
    -0.36
    deburg
    -0.35
     thrown
    -0.35
     throws
    -0.35
     threw
    -0.35
    POSITIVE LOGITS
     Southland
    0.47
    الحياه
    0.45
    outSlope
    0.44
    skraft
    0.43
     تكبرها
    0.43
     EconPapers
    0.42
     برانيه
    0.41
     devriez
    0.41
    ├──
    0.39
    Portale
    0.39
    Activations Density 0.375%

    No Known Activations