© 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 EvalsExportsinterp-engineNEWAPI 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. 16704
    Prev
    Next
    INDEX
    Explanations

    proper nouns, phrases about technology mishaps, official titles, and phrases related to government investigations

    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
    362
    +0.11
    0.5%
    207
    +0.11
    0.5%
    712
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.04
    207
    +0.11
    0.03
    656
    +0.10
    0.03
    Negative Logits
     worshipped
    -0.73
    nai
    -0.68
     disliked
    -0.68
    olor
    -0.65
    ampires
    -0.64
    namese
    -0.64
    ãĥĺãĥ©
    -0.62
    utenant
    -0.62
    terday
    -0.61
    ournal
    -0.61
    POSITIVE LOGITS
     prism
    1.09
     channels
    1.08
     intermediary
    1.02
     lens
    0.97
     intermedi
    0.93
     mechanisms
    0.92
     mediation
    0.85
     backdoor
    0.80
     proxies
    0.78
     guise
    0.77
    Activations Density 0.223%

    No Known Activations