© 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. 24226
    Prev
    Next
    INDEX
    Explanations

    commas and percentages in numerical contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    67
    +0.11
    0.5%
    127
    +0.10
    0.5%
    627
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    164
    +0.11
    0.04
    127
    +0.10
    0.03
    289
    +0.10
    0.01
    Negative Logits
     guiName
    -1.12
    etheless
    -1.00
    agic
    -0.81
    [/
    -0.77
    munition
    -0.75
    borgh
    -0.74
     [/
    -0.74
    uggest
    -0.71
    wana
    -0.71
    iqueness
    -0.71
    POSITIVE LOGITS
    )"
    1.54
    ...)
    1.44
    ?)
    1.43
    ),"
    1.42
    )",
    1.33
    âĢ¦)
    1.33
     ?)
    1.32
    )
    1.31
    *)
    1.29
    ?),
    1.29
    Activations Density 0.140%

    No Known Activations