© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 4026
    Prev
    Next
    INDEX
    Explanations

    words related to philosophical and ethical discussions, including topics like morality, religion, and political ideologies

    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
    481
    +0.10
    0.5%
    612
    +0.10
    0.5%
    55
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1
    +0.10
    0.04
    389
    +0.10
    0.03
    276
    +0.10
    0.03
    Negative Logits
    racuse
    -0.80
    ebus
    -0.76
    oug
    -0.71
    assemb
    -0.69
    gran
    -0.68
    tch
    -0.67
    selection
    -0.67
    ichen
    -0.67
    ighth
    -0.65
    erk
    -0.65
    POSITIVE LOGITS
     factors
    1.15
     extremes
    0.98
     culmin
    0.93
     combined
    0.90
     contributes
    0.88
     contrasts
    0.88
     considerations
    0.87
     together
    0.87
     notwithstanding
    0.84
     points
    0.84
    Activations Density 0.119%

    No Known Activations