© 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. 1514
    Prev
    Next
    INDEX
    Explanations

    references to physical actions and interactions

    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
    186
    +0.11
    0.5%
    445
    +0.09
    0.4%
    672
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    598
    +0.11
    0.05
    283
    +0.09
    0.02
    445
    +0.08
    0.05
    Negative Logits
    isable
    -0.76
    inguishable
    -0.69
    ocative
    -0.68
    izable
    -0.66
    anuts
    -0.63
    itutes
    -0.62
    CRIPTION
    -0.61
    DX
    -0.61
    onomy
    -0.60
    etrical
    -0.60
    POSITIVE LOGITS
     alas
    1.16
     refused
    0.97
     failed
    0.95
     opted
    0.93
     chose
    0.91
    failed
    0.90
     thwarted
    0.90
     instead
    0.88
     Instead
    0.87
     hesitated
    0.86
    Activations Density 0.623%

    No Known Activations