© 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 3072
    9. 8-RES_FS3072-JB
    11. 2996
    Prev
    Next
    INDEX
    Explanations

    names of people or places, particularly "Ek" or variations containing it

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    496
    +0.15
    0.7%
    101
    +0.11
    0.5%
    721
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.12
    232
    +0.11
    0.08
    326
    +0.10
    0.09
    Negative Logits
    tenance
    -0.78
    ãĤ´ãĥ³
    -0.76
    ificent
    -0.70
    acca
    -0.69
    roads
    -0.69
    apache
    -0.69
    creen
    -0.64
    lust
    -0.64
    ################
    -0.64
    isations
    -0.63
    POSITIVE LOGITS
    ilon
    0.93
    izabeth
    0.86
    olon
    0.83
    pecially
    0.83
    uate
    0.82
    LECT
    0.82
    cellent
    0.78
    gypt
    0.77
    entric
    0.77
    lectic
    0.76
    Activations Density 1.296%

    No Known Activations