© 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. 7211
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a degree of certainty or emphasis

    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
    508
    +0.12
    0.5%
    730
    +0.09
    0.4%
    151
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    508
    +0.12
    0.02
    716
    +0.09
    0.01
    676
    +0.09
    0.01
    Negative Logits
    ocene
    -0.77
    iens
    -0.77
     Delivery
    -0.75
    iment
    -0.74
    iments
    -0.72
    adelphia
    -0.71
    HAEL
    -0.71
    aden
    -0.70
    STER
    -0.69
    VL
    -0.69
    POSITIVE LOGITS
     tame
    0.90
     straightforward
    0.89
     accurate
    0.88
     priced
    0.86
     innocuous
    0.85
     sized
    0.84
     speaking
    0.82
     compensated
    0.82
     unrem
    0.81
     harmless
    0.80
    Activations Density 0.016%

    No Known Activations