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 24576
    5. 8-RES_FS24576-JB
    6. 23468
    Prev
    Next
    INDEX
    Explanations

    expressions of personal experience and strong opinions

    oai_token-act-pair · gpt-4o-miniTriggered 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
    566
    +0.11
    0.5%
    107
    +0.09
    0.4%
    373
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.11
    0.04
    685
    +0.09
    0.02
    762
    +0.09
    0.03
    Negative Logits
    gravity
    -0.66
    flix
    -0.66
     srfAttach
    -0.63
     Mumbai
    -0.61
    uish
    -0.59
     Hau
    -0.58
     Voyager
    -0.57
    inks
    -0.56
    ospels
    -0.56
     IMAGES
    -0.56
    POSITIVE LOGITS
     ever
    1.28
     EVER
    1.23
    ever
    0.99
     imaginable
    0.94
     encountered
    0.88
     encount
    0.81
     Ever
    0.75
    Ever
    0.75
     feas
    0.72
     experien
    0.70
    Activations Density 0.084%

    No Known Activations