© 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. 9313
    Prev
    Next
    INDEX
    Explanations

    images and visual descriptions in texts

    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
    274
    +0.11
    0.5%
    483
    +0.10
    0.5%
    611
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.11
    0.04
    483
    +0.10
    0.03
    75
    +0.10
    0.03
    Negative Logits
    pton
    -0.75
    youtube
    -0.75
    EEE
    -0.73
    tymology
    -0.69
    iche
    -0.69
    bury
    -0.68
    aceutical
    -0.68
    VEN
    -0.66
    ilings
    -0.65
    rete
    -0.64
    POSITIVE LOGITS
    ered
    0.99
    rooms
    0.92
     prominently
    0.90
    biz
    0.90
    runners
    0.89
    downs
    0.86
     signs
    0.85
     how
    0.82
     similarities
    0.81
    room
    0.79
    Activations Density 0.055%

    No Known Activations