© 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 EvalsExportsAPI 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. 17668
    Prev
    Next
    INDEX
    Explanations

    information related to physical injuries, medical conditions, and incidents

    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
    208
    +0.10
    0.4%
    549
    +0.09
    0.4%
    294
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    598
    +0.10
    0.05
    208
    +0.09
    0.05
    203
    +0.09
    0.05
    Negative Logits
    roy
    -0.73
    tnc
    -0.70
    obyl
    -0.69
    edu
    -0.67
    uto
    -0.67
    iddle
    -0.67
    aring
    -0.67
    ermanent
    -0.66
    urg
    -0.66
    arity
    -0.64
    POSITIVE LOGITS
     alas
    1.08
     withdrew
    1.01
     luckily
    1.01
     fortunately
    0.99
     declined
    0.97
     retains
    0.96
     lacked
    0.92
     nevertheless
    0.92
     opted
    0.90
     thankfully
    0.90
    Activations Density 0.143%

    No Known Activations