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. 20-GEMMASCOPE-RES-131K
    6. 8220
    Prev
    Next
    INDEX
    Explanations

    terms related to traffic and signaling devices

    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_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    SuspendLayout
    -0.43
    makeConstraints
    -0.42
    écoute
    -0.42
    licité
    -0.41
    RTSC
    -0.40
    цтва
    -0.40
     transactional
    -0.40
    principalColumn
    -0.39
    aneity
    -0.39
     machete
    -0.39
    POSITIVE LOGITS
     lighthouse
    0.98
     Lighthouse
    0.74
     phare
    0.66
    +#+#
    0.61
    ighthouse
    0.59
     beacon
    0.55
     tower
    0.55
    CloseOperation
    0.51
    ligh
    0.49
    tower
    0.48
    Activations Density 0.007%

    No Known Activations