© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1541
    Prev
    Next
    INDEX
    Explanations

    steps or instructions listed in numerical order

    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
    731
    +0.13
    0.6%
    314
    +0.10
    0.5%
    692
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    731
    +0.13
    0.03
    692
    +0.10
    0.03
    482
    +0.10
    0.03
    Negative Logits
    awaru
    -0.69
     sway
    -0.65
    allowed
    -0.65
    scenes
    -0.65
     manag
    -0.64
    wagen
    -0.63
    leading
    -0.62
    espie
    -0.61
    apon
    -0.60
    hedon
    -0.60
    POSITIVE LOGITS
    Password
    1.30
    st
    1.15
     Corinthians
    1.11
    125
    1.03
    120
    1.00
    123
    0.97
    000000
    0.95
    ½
    0.93
    128
    0.89
    ST
    0.82
    Activations Density 0.047%

    No Known Activations