© 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 24576
    9. 8-RES_FS24576-JB
    11. 20511
    Prev
    Next
    INDEX
    Explanations

    statements related to political decisions and events, especially those involving legal agreements or resignations

    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
    150
    +0.10
    0.4%
    680
    +0.10
    0.4%
    98
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.05
    184
    +0.10
    0.03
    603
    +0.10
    0.04
    Negative Logits
    é¾įåĸļ士
    -0.68
    til
    -0.67
    creation
    -0.67
    bilt
    -0.67
    _>
    -0.65
    ertodd
    -0.65
    iner
    -0.64
    respective
    -0.64
     âĶľâĶĢâĶĢ
    -0.63
    omorph
    -0.62
    POSITIVE LOGITS
     unexpectedly
    0.90
     theirs
    0.88
     hers
    0.72
     abruptly
    0.71
     prematurely
    0.69
     puberty
    0.69
     him
    0.66
     unilaterally
    0.63
     them
    0.61
     mysteriously
    0.59
    Activations Density 0.338%

    No Known Activations