© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 2899
    Prev
    Next
    INDEX
    Explanations

    contradictory statements about reality and perceptions

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

    preceding "really" or "actually"

    np_acts-logits-general · gemini-2.0-flash

    actual vs perceived

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
    AndEndTag
    -0.49
    +#+
    -0.44
    +#+#
    -0.44
    SharedCtor
    -0.39
     HasFactory
    -0.38
     suspensão
    -0.37
     ResponseEntity
    -0.37
    conversions
    -0.37
     avoient
    -0.37
    Personensuche
    -0.36
    POSITIVE LOGITS
     réalité
    0.61
     reality
    0.60
    reality
    0.58
     Reality
    0.57
     actually
    0.57
    Reality
    0.57
     actuality
    0.54
     sebenarnya
    0.53
    actually
    0.52
    其實
    0.50
    Activations Density 0.390%

    No Known Activations