© 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. 7158
    Prev
    Next
    INDEX
    Explanations

    the name "Lamar" occurring in various contexts, possibly related to politics or sports events

    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
    116
    +0.12
    0.6%
    195
    +0.12
    0.5%
    55
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.01
    200
    +0.12
    0.01
    116
    +0.11
    0.01
    Negative Logits
    awei
    -0.77
    illing
    -0.76
    lasses
    -0.76
    umbnails
    -0.70
     Syri
    -0.67
    ersen
    -0.66
    moon
    -0.66
    shock
    -0.66
    swing
    -0.66
    atchewan
    -0.65
    POSITIVE LOGITS
     Lamar
    0.94
     Zam
    0.78
    ques
    0.77
     Lyndon
    0.75
     Issa
    0.75
     Cain
    0.75
    yssey
    0.69
    ious
    0.69
    inated
    0.68
    hyde
    0.67
    Activations Density 0.038%

    No Known Activations