© 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. 20385
    Prev
    Next
    INDEX
    Explanations

    references to online articles and news sources

    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
    246
    +0.11
    0.5%
    491
    +0.11
    0.5%
    755
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    246
    +0.11
    0.03
    491
    +0.11
    0.03
    755
    +0.10
    0.03
    Negative Logits
    ascal
    -0.72
    »Ĵ
    -0.67
    IDS
    -0.65
    unta
    -0.64
    avement
    -0.64
     towed
    -0.63
    opard
    -0.63
    Whoever
    -0.62
    bably
    -0.61
    cember
    -0.59
    POSITIVE LOGITS
     aloud
    1.00
    Write
    0.91
    ership
    0.90
    dress
    0.88
    ers
    0.84
    just
    0.80
    andowski
    0.80
     excerpts
    0.78
    iness
    0.76
    ying
    0.76
    Activations Density 0.032%

    No Known Activations