© 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. 12080
    Prev
    Next
    INDEX
    Explanations

    details related to daily life activities and situations

    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
    640
    +0.15
    0.7%
    454
    +0.12
    0.6%
    6
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.15
    0.16
    454
    +0.12
    0.13
    679
    +0.12
    0.13
    Negative Logits
    itionally
    -0.76
    atical
    -0.75
     advoc
    -0.73
     intermediate
    -0.71
    izoph
    -0.71
     separation
    -0.71
    iliated
    -0.70
     organis
    -0.69
     intending
    -0.69
     enrol
    -0.68
    POSITIVE LOGITS
     But
    1.18
     Meanwhile
    1.13
     Alas
    1.08
     Yet
    1.07
     Whatever
    1.05
     Likewise
    1.04
     Worse
    1.04
     Anyway
    1.02
     Thankfully
    1.02
     Fortunately
    1.01
    Activations Density 0.660%

    No Known Activations