© 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 49152
    9. 8-RES_FS49152-JB
    11. 41618
    Prev
    Next
    INDEX
    Explanations

    references to the ambiance or environment of various events and places

    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
    447
    +0.18
    0.8%
    288
    +0.11
    0.5%
    649
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.18
    0.01
    387
    +0.11
    0.01
    644
    +0.10
    0.01
    Negative Logits
    ãĥīãĥ©
    -0.71
    inker
    -0.68
    iasis
    -0.67
    arations
    -0.66
    icus
    -0.66
    inois
    -0.65
    missions
    -0.64
    gments
    -0.64
     amput
    -0.63
    entin
    -0.62
    POSITIVE LOGITS
     atmosphere
    1.06
     atmosp
    1.03
     perv
    0.96
     surrounding
    0.92
     conducive
    0.90
     Atmosp
    0.89
     backstage
    0.82
    mble
    0.81
     vibe
    0.80
     greets
    0.80
    Activations Density 0.018%

    No Known Activations