© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 6089
    Prev
    Next
    INDEX
    Explanations

    textual formatting elements and terms related to programming

    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
    447
    +0.16
    0.7%
    529
    +0.10
    0.5%
    59
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.16
    0.06
    529
    +0.10
    0.05
    761
    +0.10
    0.05
    Negative Logits
    abet
    -0.78
    justice
    -0.76
    viol
    -0.73
    iya
    -0.73
    hat
    -0.73
    irit
    -0.71
    flies
    -0.70
    bank
    -0.70
    successful
    -0.70
    yssey
    -0.70
    POSITIVE LOGITS
     optional
    0.93
     automatic
    0.89
     compulsory
    0.89
     mandatory
    0.84
     interchangeable
    0.84
     custom
    0.82
     standardized
    0.82
     redundant
    0.80
     gimm
    0.80
     graphical
    0.79
    Activations Density 0.512%

    No Known Activations