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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3-GEMMASCOPE-RES-16K
    6. 9358
    Prev
    Next
    INDEX
    Explanations

    instances of the word "fine" used in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Anfitrión
    -0.87
     autorytatywna
    -0.86
    Cyfeiriadau
    -0.78
     وتسجيلات
    -0.77
    rrggbb
    -0.74
     tartalomajánló
    -0.73
    Personendaten
    -0.71
     Henrique
    -0.70
     INTERESAR
    -0.70
    𝟹
    -0.70
    POSITIVE LOGITS
     fine
    3.41
    Fine
    3.25
    fine
    3.21
     Fine
    3.16
     FINE
    3.00
    FINE
    2.83
     fines
    1.89
     finer
    1.84
    fines
    1.72
     Fines
    1.71
    Activations Density 0.031%

    No Known Activations