© 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. 601
    Prev
    Next
    INDEX
    Explanations

    locations or directions mentioned in a text

    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
    50
    +0.13
    0.6%
    289
    +0.11
    0.5%
    468
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    50
    +0.13
    0.05
    468
    +0.11
    0.03
    289
    +0.09
    0.02
    Negative Logits
    andise
    -0.81
    è¦ļéĨĴ
    -0.69
    issance
    -0.67
    glomer
    -0.67
    riott
    -0.67
    antz
    -0.66
    cffff
    -0.65
    idated
    -0.63
    wcsstore
    -0.63
    %]
    -0.63
    POSITIVE LOGITS
    overs
    1.40
    wing
    1.12
    ward
    1.05
     fielder
    0.94
    hander
    0.94
     flank
    0.92
     wing
    0.89
    wich
    0.89
    over
    0.85
     handed
    0.85
    Activations Density 0.065%

    No Known Activations