© 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. 9324
    Prev
    Next
    INDEX
    Explanations

    instances where it's too late for something

    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
    96
    +0.11
    0.5%
    151
    +0.09
    0.4%
    70
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    96
    +0.11
    0.01
    170
    +0.09
    0.01
    178
    +0.09
    0.01
    Negative Logits
    panel
    -0.77
    iliary
    -0.72
    İĭ
    -0.66
    similar
    -0.66
    urger
    -0.66
    anooga
    -0.65
    ©¶æ
    -0.64
    paralle
    -0.64
    groups
    -0.64
    Tang
    -0.63
    POSITIVE LOGITS
     reckoning
    0.77
     Delay
    0.72
    uberty
    0.69
     anymore
    0.68
     folly
    0.64
     mistakes
    0.64
    én
    0.64
    isSpecialOrderable
    0.63
     arrival
    0.63
    thal
    0.62
    Activations Density 0.024%

    No Known Activations