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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 608
    Prev
    Next
    INDEX
    Explanations

    mentions or discussions about specific topics or issues

    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
    155
    +0.11
    0.5%
    270
    +0.10
    0.5%
    723
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.11
    0.09
    723
    +0.10
    0.08
    362
    +0.10
    0.10
    Negative Logits
    OGR
    -0.93
    SS
    -0.76
    ©¶æ¥µ
    -0.75
    quez
    -0.72
    kie
    -0.71
    hesis
    -0.71
    cue
    -0.71
    rift
    -0.70
    kai
    -0.70
    à¥
    -0.70
    POSITIVE LOGITS
     how
    1.13
     whether
    0.95
     why
    0.92
     reforming
    0.91
     topics
    0.90
     overcoming
    0.90
     improving
    0.88
     preserving
    0.83
     wanting
    0.81
     issues
    0.78
    Activations Density 1.749%

    No Known Activations