© 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. 10488
    Prev
    Next
    INDEX
    Explanations

    the word 'and'

    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
    481
    +0.17
    0.8%
    393
    +0.10
    0.4%
    713
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    713
    +0.17
    0.06
    128
    +0.10
    0.06
    393
    +0.09
    0.07
    Negative Logits
    Times
    -0.72
    busters
    -0.66
    emi
    -0.66
    iatus
    -0.66
    dog
    -0.63
    onica
    -0.63
    jit
    -0.61
    otto
    -0.61
    gest
    -0.60
    spot
    -0.60
    POSITIVE LOGITS
     furthermore
    1.36
     consequently
    1.33
     secondly
    1.24
     moreover
    1.20
     hence
    1.18
     thence
    1.16
     therefore
    1.16
     thus
    1.13
     preferably
    1.13
     thereby
    1.07
    Activations Density 0.334%

    No Known Activations