© 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. 15239
    Prev
    Next
    INDEX
    Explanations

    elements related to transition metals and their compounds in chemical reactions

    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
    ंदीखरीदारी
    -0.41
     Membrane
    -0.39
    mang
    -0.38
    getKeyCode
    -0.36
     MongoClient
    -0.36
    muri
    -0.35
     cương
    -0.35
     Maggie
    -0.35
    caman
    -0.35
     Morning
    -0.34
    POSITIVE LOGITS
     meta
    2.89
     Meta
    2.81
    Meta
    2.69
    meta
    2.69
     Met
    2.53
     MET
    2.53
     méta
    2.47
    Met
    2.42
     мета
    2.39
    メタ
    2.28
    Activations Density 1.231%

    No Known Activations