© 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. 9474
    Prev
    Next
    INDEX
    Explanations

    specific statistics or numerical values

    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
    481
    +0.18
    0.8%
    393
    +0.12
    0.6%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.18
    0.04
    684
    +0.12
    0.03
    647
    +0.10
    0.04
    Negative Logits
    iciary
    -0.78
    atro
    -0.68
    oses
    -0.67
    andem
    -0.67
    arted
    -0.67
    alogue
    -0.63
    gur
    -0.63
    arak
    -0.62
    apult
    -0.61
    Synopsis
    -0.61
    POSITIVE LOGITS
     although
    1.04
     whereas
    0.87
     despite
    0.86
     "[
    0.85
     there
    0.82
     respondents
    0.79
     contrary
    0.74
     unlike
    0.73
     whilst
    0.73
     they
    0.72
    Activations Density 0.168%

    No Known Activations