© 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. 2828
    Prev
    Next
    INDEX
    Explanations

    explicit sexual content

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    1757
    +0.08
    0.3%
    1404
    +0.07
    0.2%
    1506
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1196
    +0.08
    0.02
    882
    +0.07
    0.02
    1363
    +0.07
    0.02
    Negative Logits
    <bos>
    -1.09
    ുറ
    -0.67
    public
    -0.65
    +#+#
    -0.64
    ⓧ
    -0.64
    Portály
    -0.60
    /**
    -0.60
    <?
    -0.60
    place
    -0.59
    //};
    -0.59
    POSITIVE LOGITS
     fuck
    1.59
     fucks
    1.46
     FUCK
    1.33
     Juf
    1.30
     bandung
    1.29
     véhic
    1.28
     Minang
    1.28
     Fuck
    1.26
     maneu
    1.25
     scrat
    1.25
    Activations Density 0.239%

    No Known Activations