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. 20-GEMMASCOPE-RES-16K
    6. 4149
    Prev
    Next
    INDEX
    Explanations

    phrases related to lack of time or excuses for not taking action

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Followed by verbs and personal pronouns

    np_acts-logits-general · gemini-2.0-flash

    future possibility or sufficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    SharedDtor
    -0.89
    RegressionTest
    -0.81
    脚注の使い方
    -0.78
    twimg
    -0.76
     ProtoMessage
    -0.75
    TagMode
    -0.72
    InputTagHelper
    -0.71
     BoxFit
    -0.71
    __":
    -0.71
    verifyException
    -0.70
    POSITIVE LOGITS
     dare
    0.85
     dared
    0.75
     dares
    0.73
     daring
    0.71
     enough
    0.71
     available
    0.69
     courage
    0.68
     brave
    0.68
    勇
    0.65
     lazy
    0.64
    Activations Density 0.294%

    No Known Activations