Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 631
    Prev
    Next
    INDEX
    Explanations

    descriptions and discussions related to various subjects with detailed points and comparisons

    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
    408
    +0.11
    0.5%
    206
    +0.10
    0.5%
    608
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.11
    0.19
    362
    +0.10
    0.17
    616
    +0.10
    0.10
    Negative Logits
    byter
    -0.72
    Sov
    -0.72
    bow
    -0.71
    bus
    -0.71
    ope
    -0.70
    otti
    -0.70
    swick
    -0.69
    nesty
    -0.68
    iversity
    -0.68
    20439
    -0.66
    POSITIVE LOGITS
     these
    0.95
     modern
    0.90
     this
    0.89
     humankind
    0.85
     mankind
    0.82
     our
    0.81
     contemporary
    0.80
     everyday
    0.79
     constructing
    0.77
     each
    0.77
    Activations Density 3.834%

    No Known Activations