Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 23-GEMMASCOPE-RES-131K
    6. 99
    Prev
    Next
    INDEX
    Explanations

    mathematical operations and expressions, particularly derivatives and definitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Hook Layer
    23
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Földrajzportál
    -0.35
     rather
    -0.35
     přip
    -0.35
     היתה
    -0.33
    意义
    -0.32
     muchacha
    -0.32
     bulundu
    -0.31
     magique
    -0.31
     czyli
    -0.30
     guerrero
    -0.30
    POSITIVE LOGITS
    RegressionTest
    0.97
    ロウィン
    0.66
    <unused41>
    0.65
    <unused17>
    0.65
    <unused79>
    0.65
    <unused23>
    0.65
    <unused42>
    0.65
    <unused14>
    0.65
    [@BOS@]
    0.65
    <unused8>
    0.65
    Activations Density 0.012%

    No Known Activations