© 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 12288
    9. 8-RES_FS12288-JB
    11. 401
    Prev
    Next
    INDEX
    Explanations

    statements about concerns, statistics, and research findings

    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
    120
    +0.11
    0.5%
    373
    +0.11
    0.5%
    410
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.11
    0.04
    566
    +0.11
    0.07
    157
    +0.10
    0.07
    Negative Logits
    Fuck
    -0.76
    fuck
    -0.74
     fuck
    -0.71
    die
    -0.67
     dude
    -0.64
     crap
    -0.64
     decency
    -0.62
     goddamn
    -0.62
     Fuck
    -0.62
     godd
    -0.61
    POSITIVE LOGITS
     therefore
    1.36
     furthermore
    1.36
     however
    1.29
     moreover
    1.26
     also
    1.25
     meanwhile
    1.22
     additionally
    1.20
     thus
    1.06
    also
    0.86
     meantime
    0.86
    Activations Density 0.873%

    No Known Activations