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

    Neuronpedia

    Jacobian LensNEW
    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. 17422
    Prev
    Next
    INDEX
    Explanations

    reviews or opinions expressing preferences or criticisms

    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
    78
    +0.09
    0.4%
    608
    +0.09
    0.4%
    563
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.09
    0.04
    766
    +0.09
    0.03
    563
    +0.09
    0.03
    Negative Logits
    Cause
    -0.71
    COL
    -0.70
    scl
    -0.67
    PATH
    -0.67
    kr
    -0.66
    Interstitial
    -0.66
    IGHTS
    -0.65
     actionGroup
    -0.65
    anners
    -0.65
    ONEY
    -0.65
    POSITIVE LOGITS
     this
    0.85
     these
    0.81
     nutshell
    0.71
     deploying
    0.66
     modern
    0.65
     researching
    0.65
     constructing
    0.63
     adopting
    0.61
     blogging
    0.61
     staying
    0.61
    Activations Density 0.149%

    No Known Activations