© 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. 45605
    Prev
    Next
    INDEX
    Explanations

    references to specific individuals in context

    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
    484
    +0.11
    0.5%
    710
    +0.10
    0.5%
    420
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.11
    0.01
    659
    +0.10
    0.01
    406
    +0.10
    0.01
    Negative Logits
    anmar
    -0.80
    ifted
    -0.76
    arcity
    -0.76
     glim
    -0.76
    ifies
    -0.75
    uitous
    -0.74
    ENCY
    -0.74
    ifying
    -0.73
    ific
    -0.73
    committee
    -0.73
    POSITIVE LOGITS
    lla
    0.93
    ette
    0.89
    que
    0.85
    ttes
    0.83
    lli
    0.80
    vre
    0.80
    llo
    0.79
    brate
    0.77
     Hebdo
    0.76
    brates
    0.76
    Activations Density 0.008%

    No Known Activations