© 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. 26655
    Prev
    Next
    INDEX
    Explanations

    references to television talk shows

    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
    679
    +0.11
    0.5%
    414
    +0.10
    0.5%
    151
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    76
    +0.11
    0.02
    115
    +0.10
    0.02
    151
    +0.10
    0.02
    Negative Logits
    quez
    -0.73
    illard
    -0.71
    ©¶æ¥µ
    -0.70
    eco
    -0.68
    FIELD
    -0.68
    lda
    -0.67
    etsy
    -0.67
     scarce
    -0.66
    cit
    -0.65
    rez
    -0.65
    POSITIVE LOGITS
    runners
    1.12
    biz
    1.12
     hosts
    1.09
     host
    1.03
    runner
    1.02
     contestant
    0.97
     presenter
    0.95
     mogul
    0.90
    case
    0.87
     CNBC
    0.86
    Activations Density 0.025%

    No Known Activations