© 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. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 10-RES-JB
    11. 4930
    Prev
    Next
    INDEX
    Explanations

    instances of the word "talk" in various forms

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.24
    1.3%
    538
    +0.13
    0.7%
    1296
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    538
    +0.24
    0.04
    1296
    +0.13
    0.04
    1806
    +0.12
    0.04
    Negative Logits
    <bos>
    -2.70
    ///**
    -0.66
    /***
    -0.66
     pel
    -0.60
     per
    -0.56
    import
    -0.55
    ////
    -0.55
     pen
    -0.54
    dom
    -0.54
    代
    -0.53
    POSITIVE LOGITS
     Minang
    1.44
     pymysql
    1.43
     sovere
    1.37
     Juf
    1.36
     unwarran
    1.36
     maneu
    1.33
     Khart
    1.33
     stockholm
    1.31
     Sarm
    1.28
     bandung
    1.27
    Activations Density 0.045%

    No Known Activations