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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 10597
    Prev
    Next
    INDEX
    Explanations

    technical terminology and concepts related to scientific experiments and data analysis

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

    content related to data processing, corrections, or modifications in scientific and technical contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    scientific-technical language denoting processed or transformed states, especially past participles and -tion/-ed morphological forms describing data or material modifications.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    words ending in "-ed" or related to past participles and verb modifications.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
    Hentet
    -0.78
    ſehen
    -0.74
     utafitiHapana
    -0.73
    expandindo
    -0.67
     للمعارف
    -0.67
    AddTagHelper
    -0.66
     ddelwed
    -0.66
     témoig
    -0.65
    <unused8>
    -0.65
    <unused3>
    -0.65
    POSITIVE LOGITS
     "..\..\
    0.31
     makes
    0.28
    AW
    0.27
    makes
    0.26
     الو
    0.26
     tun
    0.24
    Locus
    0.24
    Makes
    0.24
     Makes
    0.23
     waardoor
    0.23
    Activations Density 0.699%

    No Known Activations