© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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 12288
    9. 8-RES_FS12288-JB
    11. 1929
    Prev
    Next
    INDEX
    Explanations

    phrases related to danger or threat to lives

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    759
    +0.14
    0.6%
    132
    +0.13
    0.6%
    152
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    132
    +0.14
    0.04
    759
    +0.13
    0.04
    575
    +0.10
    0.03
    Negative Logits
    xual
    -0.92
    RECT
    -0.71
    ority
    -0.71
    OSP
    -0.69
    Applic
    -0.66
    NES
    -0.65
    VIS
    -0.65
    Null
    -0.63
    Stud
    -0.63
    atchewan
    -0.63
    POSITIVE LOGITS
    lihood
    0.90
    mares
    0.90
     endanger
    0.84
    guards
    0.83
     endangered
    0.82
    guard
    0.78
    journal
    0.76
    cape
    0.75
    warming
    0.75
    vas
    0.75
    Activations Density 0.086%

    No Known Activations