© 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. 10407
    Prev
    Next
    INDEX
    Explanations

    terms related to hostile actions or threats, particularly hijacking or terrorist activities

    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
    +0.15
    0.7%
    526
    +0.13
    0.6%
    733
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    733
    +0.15
    0.01
    163
    +0.13
    0.01
    403
    +0.12
    0.01
    Negative Logits
    ç¥ŀ
    -0.72
     Rite
    -0.70
    20439
    -0.69
     Angels
    -0.69
     Solitaire
    -0.69
     Staples
    -0.68
     Pompe
    -0.66
     Thick
    -0.63
     Balt
    -0.62
     Vlad
    -0.62
    POSITIVE LOGITS
    ackers
    1.14
    acker
    1.08
    acking
    1.05
     hij
    1.00
    inx
    0.98
    intosh
    0.90
    adish
    0.85
    abis
    0.82
    agic
    0.79
    acked
    0.78
    Activations Density 0.020%

    No Known Activations