© 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. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 0-RES-JB
    11. 2395
    Prev
    Next
    INDEX
    Explanations

    instances of the word "standing"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.10
    0.4%
    1573
    +0.05
    0.2%
    9
    +0.05
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1076
    +0.10
    0.04
    1370
    +0.05
    0.03
    443
    +0.05
    0.03
    Negative Logits
    <bos>
    -1.62
    ⓧ
    -0.68
    public
    -0.65
    WriteLiteral
    -0.64
    <?
    -0.62
    //
    -0.62
    Enllaços
    -0.62
    }{||
    -0.61
    ///**
    -0.60
    ുറ
    -0.60
    POSITIVE LOGITS
     standing
    2.03
     Standing
    1.97
    Standing
    1.81
    standing
    1.57
     stockholm
    1.51
     STAND
    1.38
     applau
    1.38
     Juf
    1.38
     fta
    1.35
     mef
    1.35
    Activations Density 0.064%

    No Known Activations