© 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. 43320
    Prev
    Next
    INDEX
    Explanations

    references to characters and concepts from the Star Wars franchise

    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
    441
    +0.11
    0.5%
    59
    +0.10
    0.5%
    549
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.01
    326
    +0.10
    0.00
    588
    +0.10
    0.01
    Negative Logits
     htt
    -0.84
    uckland
    -0.77
    uters
    -0.74
    mingham
    -0.72
    case
    -0.69
    ropri
    -0.68
    trak
    -0.68
    uments
    -0.68
    ãĥ¼ãĥ³
    -0.67
    anwhile
    -0.67
    POSITIVE LOGITS
     Vader
    1.61
     Maul
    1.16
     Bane
    0.97
    rome
    0.88
    rell
    0.87
    lord
    0.84
    orian
    0.80
     Plague
    0.78
     Sid
    0.75
     Darth
    0.74
    Activations Density 0.004%

    No Known Activations