© 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. 18613
    Prev
    Next
    INDEX
    Explanations

    the word "way" with varying activation levels

    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
    737
    +0.13
    0.6%
    240
    +0.10
    0.5%
    50
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    737
    +0.13
    0.03
    435
    +0.10
    0.02
    50
    +0.10
    0.02
    Negative Logits
    iasco
    -0.73
    lict
    -0.72
    uster
    -0.72
     livest
    -0.70
    usters
    -0.69
    uctor
    -0.64
    iners
    -0.64
    uates
    -0.63
     fumes
    -0.62
    ividual
    -0.62
    POSITIVE LOGITS
    fare
    1.18
    ward
    1.18
    finding
    1.05
    points
    1.04
    point
    1.03
    forward
    0.96
    bill
    0.93
    WARD
    0.91
    Forward
    0.89
    cross
    0.89
    Activations Density 0.029%

    No Known Activations