© 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 1536
    9. 8-RES_FS1536-JB
    11. 47
    Prev
    Next
    INDEX
    Explanations

    terms related to technology, development, and industry

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    288
    +0.14
    0.7%
    376
    +0.11
    0.5%
    481
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.19
    575
    +0.11
    0.14
    724
    +0.11
    0.14
    Negative Logits
    tein
    -1.04
    drawn
    -0.77
    nell
    -0.73
    ibo
    -0.69
    ihad
    -0.69
     Lama
    -0.65
    ippi
    -0.65
     Duchess
    -0.64
    cop
    -0.63
    nance
    -0.63
    POSITIVE LOGITS
    wide
    0.99
     revolves
    0.87
    eers
    0.86
    Wide
    0.81
     evolves
    0.80
     rife
    0.80
    places
    0.79
     sclerosis
    0.77
     dominated
    0.76
    atics
    0.76
    Activations Density 2.709%

    No Known Activations