© 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 49152
    9. 8-RES_FS49152-JB
    11. 41541
    Prev
    Next
    INDEX
    Explanations

    instances of the word "just."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    633
    +0.12
    0.5%
    508
    +0.10
    0.5%
    106
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    633
    +0.12
    0.04
    106
    +0.10
    0.03
    498
    +0.10
    0.03
    Negative Logits
    ccording
    -0.69
     Palestin
    -0.67
     adolesc
    -0.65
     Strategies
    -0.64
     subsequ
    -0.64
     Key
    -0.64
     discrep
    -0.63
     challeng
    -0.61
     Archdemon
    -0.59
     representations
    -0.59
    POSITIVE LOGITS
    ifiable
    1.15
    ifications
    1.07
     ignore
    1.03
    ify
    1.02
    if
    0.98
    itate
    0.92
    ificate
    0.89
    ification
    0.89
    ified
    0.87
    ifi
    0.85
    Activations Density 0.040%

    No Known Activations