© 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. 19816
    Prev
    Next
    INDEX
    Explanations

    profanity

    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
    526
    +0.12
    0.6%
    750
    +0.12
    0.5%
    48
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    750
    +0.12
    0.02
    48
    +0.12
    0.01
    326
    +0.11
    0.01
    Negative Logits
     rece
    -0.78
     obser
    -0.75
    PsyNetMessage
    -0.72
    BIL
    -0.70
     membr
    -0.68
     Geographic
    -0.66
     conclud
    -0.65
     reper
    -0.64
    HCR
    -0.64
     conve
    -0.63
    POSITIVE LOGITS
    holes
    1.21
    hole
    1.10
    bags
    1.04
    lord
    1.00
    nuts
    0.98
    bag
    0.97
    wit
    0.94
    ups
    0.93
    fuck
    0.92
    yeah
    0.90
    Activations Density 0.017%

    No Known Activations