© 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. 7196
    Prev
    Next
    INDEX
    Explanations

    instances of the word 'The'

    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
    480
    +0.33
    1.8%
    447
    +0.32
    1.8%
    481
    +0.20
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.33
    0.12
    385
    +0.32
    0.11
    608
    +0.20
    0.11
    Negative Logits
    Ò
    -0.63
    omever
    -0.62
    thood
    -0.62
     beforehand
    -0.62
    !.
    -0.62
    ÏĢ
    -0.60
    *.
    -0.58
    Ó
    -0.57
     alongside
    -0.57
    ãĥ¼ãĥĨ
    -0.56
    POSITIVE LOGITS
    oret
    1.14
    resa
    1.08
    odore
    1.00
    ories
    0.82
     latter
    0.81
     biggest
    0.80
     oldest
    0.77
     idea
    0.75
     majority
    0.75
     largest
    0.75
    Activations Density 0.517%

    No Known Activations