© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 0-GEMMASCOPE-RES-65K
    11. 13444
    Prev
    Next
    INDEX
    Explanations

    instances of the word "hard"

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.0.hook_resid_post
    Hook Layer
    0
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    SpringRunner
    -0.68
     pyjamas
    -0.66
     CiNii
    -0.64
     sumpay
    -0.61
    unhofer
    -0.60
    ineno
    -0.60
     Wikimédia
    -0.59
     Mercure
    -0.59
     telescopes
    -0.58
    olves
    -0.57
    POSITIVE LOGITS
     hard
    3.55
     Hard
    3.08
    Hard
    3.03
    hard
    3.03
     HARD
    2.80
    HARD
    2.63
     harder
    2.30
     tough
    2.28
     hardest
    2.27
     harde
    2.06
    Activations Density 0.047%

    No Known Activations