© 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. 0-RES-JB
    11. 2576
    Prev
    Next
    INDEX
    Explanations

    references to legal matters or contracts

    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
    +0.09
    0.3%
    596
    +0.09
    0.3%
    479
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    586
    +0.09
    0.03
    899
    +0.09
    0.03
    596
    +0.07
    0.03
    Negative Logits
    <bos>
    -1.31
    ///**
    -0.79
    /***
    -0.75
    public
    -0.72
    export
    -0.68
    addGroup
    -0.66
    exports
    -0.65
    hline
    -0.65
    void
    -0.64
    }{||
    -0.64
    POSITIVE LOGITS
     assumption
    2.06
     affor
    1.93
     maneu
    1.93
     Assume
    1.92
     Assumptions
    1.91
     assumptions
    1.91
    assume
    1.89
     assume
    1.88
    assumption
    1.87
     stockholm
    1.86
    Activations Density 0.132%

    No Known Activations