© 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-2-9B
    7. Residual Stream - 16k
    9. 28-GEMMASCOPE-RES-16K
    11. 11845
    Prev
    Next
    INDEX
    Explanations

    instances of the first-person pronoun "I" and variations of the verb "to be."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    Follows "I am" or "to"

    np_acts-logits-general · gemini-2.0-flash

    am too, am starting, to cell

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     فريبيس
    -0.71
     ſche
    -0.70
    MessageTagHelper
    -0.69
     يتيمه
    -0.68
    parsedMessage
    -0.68
     ſtate
    -0.66
    ReusableCell
    -0.65
     itſelf
    -0.65
    verwijspagina
    -0.64
     poffible
    -0.64
    POSITIVE LOGITS
     constantly
    0.52
     makhluk
    0.47
     always
    0.47
     committed
    0.43
     unwilling
    0.43
     willing
    0.42
     binatang
    0.42
     animados
    0.41
     incapable
    0.41
     capable
    0.40
    Activations Density 0.317%

    No Known Activations