© 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. 19869
    Prev
    Next
    INDEX
    Explanations

    questions related to the duration of time

    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
    60
    +0.12
    0.5%
    361
    +0.11
    0.5%
    529
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    60
    +0.12
    0.02
    529
    +0.11
    0.02
    361
    +0.10
    0.02
    Negative Logits
    nikov
    -0.77
    ħĭ
    -0.75
    unin
    -0.75
    eni
    -0.74
    eking
    -0.70
    isha
    -0.69
    illard
    -0.69
    ooter
    -0.69
    rored
    -0.68
     Cosponsors
    -0.67
    POSITIVE LOGITS
     ago
    1.24
     elapsed
    1.03
     periods
    0.99
     lasting
    0.93
    ago
    0.89
     awaited
    0.87
     duration
    0.84
     Ago
    0.82
     Term
    0.80
     waits
    0.80
    Activations Density 0.024%

    No Known Activations