© 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 1536
    9. 8-RES_FS1536-JB
    11. 1049
    Prev
    Next
    INDEX
    Explanations

    the word "way" with a high activation value

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    160
    +0.10
    0.5%
    46
    +0.10
    0.5%
    50
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    705
    +0.10
    0.07
    737
    +0.10
    0.07
    50
    +0.10
    0.07
    Negative Logits
    usters
    -1.02
    uster
    -0.88
     livest
    -0.84
    ĸļ
    -0.79
    omore
    -0.73
    oppable
    -0.71
    asts
    -0.69
    oubted
    -0.69
    inately
    -0.68
    lict
    -0.67
    POSITIVE LOGITS
    fare
    1.27
    finding
    1.21
    ward
    1.19
    forward
    1.10
    point
    1.06
    finder
    0.90
    points
    0.89
     forward
    0.88
    bill
    0.81
    station
    0.77
    Activations Density 1.161%

    No Known Activations