© 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. 1164
    Prev
    Next
    INDEX
    Explanations

    information related to practical skills and training programs, such as identifying weeds and plants, soil fertility, livestock management, veterinary skills, and coaching techniques

    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%
    1870
    +0.07
    0.3%
    2004
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    585
    +0.10
    0.04
    2004
    +0.07
    0.03
    1075
    +0.06
    0.03
    Negative Logits
    <bos>
    -2.05
    
    -0.78
    enumerate
    -0.78
    public
    -0.77
    <?
    -0.75
    /*
    -0.74
    displayquote
    -0.74
    #
    -0.74
    .
    -0.74
    var
    -0.73
    POSITIVE LOGITS
     fta
    2.50
     ftu
    2.45
     increa
    2.39
     emphat
    2.33
     fte
    2.33
     affor
    2.32
     aen
    2.31
     effe
    2.31
     unden
    2.29
     maneu
    2.27
    Activations Density 0.071%

    No Known Activations