© 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. 1827
    Prev
    Next
    INDEX
    Explanations

    the word "remains" or variations of it in different contexts

    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
    735
    +0.11
    0.5%
    642
    +0.11
    0.5%
    195
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.11
    0.03
    735
    +0.11
    0.03
    102
    +0.10
    0.04
    Negative Logits
    ramid
    -0.91
    PsyNetMessage
    -0.70
    MAT
    -0.69
    isson
    -0.68
    gee
    -0.68
    ongyang
    -0.66
    Mart
    -0.61
     Blossom
    -0.61
     externalToEVAOnly
    -0.61
    idium
    -0.60
    POSITIVE LOGITS
     unchanged
    1.32
     intact
    1.25
     unanswered
    1.20
     unsolved
    1.14
     untouched
    1.12
     unaffected
    1.03
     steadfast
    1.02
     unresolved
    1.00
     undefeated
    1.00
     undecided
    0.96
    Activations Density 0.064%

    No Known Activations