© 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 12288
    9. 8-RES_FS12288-JB
    11. 10523
    Prev
    Next
    INDEX
    Explanations

    mentions of time durations involving the word "hours."

    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
    178
    +0.11
    0.5%
    294
    +0.11
    0.5%
    651
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.11
    0.04
    657
    +0.11
    0.03
    178
    +0.10
    0.03
    Negative Logits
    dinand
    -0.91
    LECT
    -0.74
    bard
    -0.72
    reditary
    -0.72
     Sovere
    -0.68
    esan
    -0.66
    Magikarp
    -0.65
    raviolet
    -0.65
    oid
    -0.64
     Instit
    -0.64
    POSITIVE LOGITS
    glass
    1.27
     Ago
    0.82
    endment
    0.81
     commute
    0.79
    Hours
    0.76
     elapsed
    0.76
     logged
    0.75
     intervals
    0.73
     ILCS
    0.70
    long
    0.70
    Activations Density 0.064%

    No Known Activations