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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 11540
    Prev
    Next
    INDEX
    Explanations

    phrases related to preparation and readiness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    InputModule
    -0.52
     borboleta
    -0.52
     հղումներ
    -0.49
    ьаж
    -0.49
     caneta
    -0.48
    addCriterion
    -0.48
    ganggu
    -0.48
     aranha
    -0.47
    oa̍t
    -0.46
    anyeol
    -0.46
    POSITIVE LOGITS
     prepare
    1.22
     Prepare
    1.18
    prepare
    1.16
     preparation
    1.16
     prepared
    1.15
     Preparation
    1.13
    Prepare
    1.12
    prepared
    1.12
     ready
    1.11
     Ready
    1.11
    Activations Density 0.116%

    No Known Activations