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-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 7259
    Prev
    Next
    INDEX
    Explanations

    conditional statements and hypothetical scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Spitz
    -0.63
     getRule
    -0.60
     Nouveau
    -0.56
    っしゃ
    -0.56
     combineReducers
    -0.55
    hasNext
    -0.55
    NOPQRST
    -0.53
    loaf
    -0.53
    webtoken
    -0.52
     organizador
    -0.51
    POSITIVE LOGITS
     if
    1.64
     If
    1.49
    if
    1.49
    If
    1.39
     если
    1.30
     nếu
    1.29
     jika
    1.24
     hvis
    1.20
    Если
    1.20
     Если
    1.20
    Activations Density 0.328%

    No Known Activations