© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9502
    Prev
    Next
    INDEX
    Explanations

    instances of creators, owners, makers, producers, founders, directors, or executives

    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
    204
    +0.10
    0.5%
    373
    +0.10
    0.5%
    291
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    204
    +0.10
    0.04
    104
    +0.10
    0.04
    264
    +0.09
    0.03
    Negative Logits
    humans
    -0.73
    tnc
    -0.67
     Matters
    -0.64
    exist
    -0.61
     \'
    -0.60
     Definitions
    -0.59
    olars
    -0.59
    Americans
    -0.59
    Semitism
    -0.58
    uld
    -0.57
    POSITIVE LOGITS
    liest
    1.06
     overseeing
    1.03
     of
    0.95
    iest
    0.95
     responsible
    0.81
    agonist
    0.80
     thereof
    0.80
     presiding
    0.79
     incarnation
    0.73
    ultimate
    0.72
    Activations Density 0.136%

    No Known Activations