© 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
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 23879
    Prev
    Next
    INDEX
    Explanations

    references to time periods or durations

    oai_token-act-pair · gpt-4o-miniTriggered 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
    361
    +0.12
    0.5%
    633
    +0.11
    0.5%
    545
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    657
    +0.12
    0.04
    651
    +0.11
    0.04
    361
    +0.10
    0.04
    Negative Logits
     heights
    -0.77
    eus
    -0.66
     aloud
    -0.65
     Cir
    -0.60
     Rooms
    -0.59
    edia
    -0.59
     Halls
    -0.58
    ech
    -0.58
     Mysteries
    -0.57
    circle
    -0.56
    POSITIVE LOGITS
     ago
    0.83
     elapsed
    0.76
    manship
    0.76
    folio
    0.73
    wings
    0.72
     spare
    0.71
    glass
    0.71
    hift
    0.70
    ional
    0.70
     chance
    0.69
    Activations Density 0.098%

    No Known Activations