© 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 12288
    9. 8-RES_FS12288-JB
    11. 1913
    Prev
    Next
    INDEX
    Explanations

    situations where someone delays an elevator and holds the door open for a stranger

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    480
    +0.44
    2.3%
    266
    +0.11
    0.6%
    101
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    674
    +0.44
    0.05
    23
    +0.11
    0.05
    266
    +0.09
    0.07
    Negative Logits
     phenomen
    -0.53
     astron
    -0.51
     Dru
    -0.50
     Guatem
    -0.50
    Cro
    -0.49
     Hodg
    -0.49
     Hob
    -0.49
     Cro
    -0.47
     advoc
    -0.47
     Hawkins
    -0.47
    POSITIVE LOGITS
    actionDate
    0.75
    ohm
    0.74
    displayText
    0.70
    ãĥį
    0.68
    adeon
    0.63
    raphics
    0.60
    nces
    0.60
    ampire
    0.60
    OTOS
    0.60
    racuse
    0.59
    Activations Density 0.733%

    No Known Activations