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

    Neuronpedia

    Jacobian LensNEW
    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 24576
    9. 8-RES_FS24576-JB
    11. 5808
    Prev
    Next
    INDEX
    Explanations

    references to specific characters and elements from fictional universes

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    481
    +0.12
    0.6%
    678
    +0.10
    0.4%
    197
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.12
    0.04
    678
    +0.10
    0.04
    673
    +0.10
    0.04
    Negative Logits
    jri
    -0.81
    ertodd
    -0.77
    lah
    -0.74
    hap
    -0.73
    cipled
    -0.73
    reply
    -0.73
    ntil
    -0.72
    iquid
    -0.71
    orget
    -0.70
    animous
    -0.69
    POSITIVE LOGITS
     trilogy
    1.28
     onwards
    1.13
     Characters
    1.12
     Trilogy
    1.11
     sequels
    1.07
     canon
    1.06
    canon
    1.05
     Avatar
    1.05
     storyline
    1.04
     onward
    1.04
    Activations Density 0.330%

    No Known Activations