© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 41512
    Prev
    Next
    INDEX
    Explanations

    references to collective experiences and interactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    459
    +0.15
    0.7%
    641
    +0.12
    0.5%
    329
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.15
    0.04
    45
    +0.12
    0.03
    285
    +0.11
    0.03
    Negative Logits
     disappro
    -0.70
    ails
    -0.63
    merce
    -0.63
    lled
    -0.63
     racket
    -0.62
    unts
    -0.62
     panicked
    -0.62
     mysteriously
    -0.61
     retracted
    -0.61
    emic
    -0.60
    POSITIVE LOGITS
     preferably
    1.09
     Ideally
    1.09
     hopefully
    1.04
     Specifically
    0.99
     ideally
    0.99
    Specifically
    0.89
    rather
    0.88
     secondly
    0.84
     Hopefully
    0.82
     Because
    0.80
    Activations Density 0.361%

    No Known Activations