© 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. 10977
    Prev
    Next
    INDEX
    Explanations

    phrases relating to relationships with partners, family dynamics, and personal struggles

    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
    59
    +0.12
    0.5%
    502
    +0.10
    0.5%
    745
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.12
    0.05
    762
    +0.10
    0.06
    15
    +0.10
    0.05
    Negative Logits
    CNN
    -0.73
    utical
    -0.72
     Bundes
    -0.70
     Period
    -0.68
     Actress
    -0.67
     sovere
    -0.66
     EW
    -0.63
    nb
    -0.63
    PLA
    -0.60
     unification
    -0.59
    POSITIVE LOGITS
     ejac
    0.97
     handsome
    0.89
     circumcised
    0.81
     semen
    0.81
     penis
    0.80
    aped
    0.78
     sperm
    0.76
     cock
    0.76
    ading
    0.75
    venants
    0.74
    Activations Density 0.252%

    No Known Activations