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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 1549
    Prev
    Next
    INDEX
    Explanations

    when the phrase "comes to mind" is used

    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
    756
    +0.15
    0.7%
    665
    +0.13
    0.6%
    373
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.15
    0.03
    568
    +0.13
    0.02
    101
    +0.12
    0.03
    Negative Logits
     Princ
    -0.65
     Auckland
    -0.63
     Plans
    -0.60
    wake
    -0.60
     Gord
    -0.58
     Lyons
    -0.58
    ulum
    -0.57
     DeL
    -0.56
     ambition
    -0.56
     Dragons
    -0.56
    POSITIVE LOGITS
     unexpectedly
    0.76
    Film
    0.70
     suddenly
    0.70
    EVA
    0.68
     smelling
    0.68
     spontaneously
    0.68
     differently
    0.67
    heet
    0.67
    ixel
    0.66
     blows
    0.65
    Activations Density 0.067%

    No Known Activations