© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 16857
    Prev
    Next
    INDEX
    Explanations

    terms related to chemical substances or hazards

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    288
    +0.11
    0.5%
    121
    +0.10
    0.5%
    516
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.11
    0.02
    121
    +0.10
    0.02
    658
    +0.10
    0.02
    Negative Logits
    doms
    -0.87
    alos
    -0.84
    olulu
    -0.77
    abouts
    -0.77
    iak
    -0.76
    alach
    -0.74
    ership
    -0.73
    church
    -0.72
     4090
    -0.71
    naire
    -0.71
    POSITIVE LOGITS
     makeup
    1.06
     compounds
    0.99
     substances
    0.98
     chemical
    0.98
     warfare
    0.97
     additives
    0.97
     chemicals
    0.94
     residue
    0.92
     oxide
    0.92
     chlorine
    0.91
    Activations Density 0.025%

    No Known Activations