© 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 24576
    9. 8-RES_FS24576-JB
    11. 9667
    Prev
    Next
    INDEX
    Explanations

    athlete names and sports terms

    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
    373
    +0.15
    0.7%
    636
    +0.10
    0.5%
    700
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.15
    0.07
    373
    +0.10
    -0.02
    232
    +0.09
    0.06
    Negative Logits
     plates
    -0.64
     psi
    -0.57
     ultras
    -0.54
     mortar
    -0.54
     coli
    -0.53
     Peninsula
    -0.53
     roadside
    -0.52
     storefront
    -0.51
     fract
    -0.51
     brill
    -0.51
    POSITIVE LOGITS
     Jr
    1.01
    chuk
    0.96
    opoulos
    0.85
    zynski
    0.80
    baugh
    0.79
    midt
    0.78
    isner
    0.77
    berger
    0.76
    iev
    0.76
    acker
    0.76
    Activations Density 0.199%

    No Known Activations