© 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. 10731
    Prev
    Next
    INDEX
    Explanations

    descriptive adjectives related to behavior and attributes

    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
    408
    +0.12
    0.6%
    258
    +0.11
    0.5%
    362
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.12
    0.06
    381
    +0.11
    0.04
    387
    +0.11
    0.04
    Negative Logits
    osponsors
    -0.93
    é¾į
    -0.88
     Parables
    -0.77
    apper
    -0.75
    channelAvailability
    -0.73
    911
    -0.64
    eper
    -0.64
     sidx
    -0.64
    conom
    -0.64
    oji
    -0.63
    POSITIVE LOGITS
     bast
    0.72
    istic
    0.65
     STD
    0.62
     appar
    0.62
     Crimean
    0.62
     affair
    0.60
    izational
    0.59
     Commonwealth
    0.59
     gameplay
    0.59
     slog
    0.59
    Activations Density 0.200%

    No Known Activations