© 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 24576
    9. 8-RES_FS24576-JB
    11. 20061
    Prev
    Next
    INDEX
    Explanations

    threatening and authoritative language in communications

    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
    351
    +0.11
    0.5%
    480
    +0.10
    0.5%
    526
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    351
    +0.11
    0.06
    526
    +0.10
    0.00
    665
    +0.10
    0.07
    Negative Logits
    quartered
    -0.93
     strikingly
    -0.78
    xtap
    -0.73
     opting
    -0.69
    atial
    -0.68
     ostensibly
    -0.68
     anecd
    -0.68
     unusually
    -0.67
     allegedly
    -0.67
     prompted
    -0.66
    POSITIVE LOGITS
     tomorrow
    1.45
     someday
    1.23
    morrow
    1.14
     fuckin
    1.09
    !"
    1.09
     whoever
    1.06
     tonight
    1.02
     ..."
    1.02
    !!"
    1.01
     âĢ¦"
    0.99
    Activations Density 0.851%

    No Known Activations