© 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 EvalsExportsinterp-engineNEWAPI 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. 15996
    Prev
    Next
    INDEX
    Explanations

    information related to statistics or progress updates

    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
    373
    +0.20
    0.9%
    181
    +0.10
    0.5%
    506
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    181
    +0.20
    0.05
    455
    +0.10
    0.04
    495
    +0.08
    0.04
    Negative Logits
     Parables
    -0.84
    ĸļ
    -0.84
    âĶģ
    -0.75
    dan
    -0.72
     Agric
    -0.72
     Paran
    -0.69
    Tes
    -0.69
    Clar
    -0.69
    ournals
    -0.67
     Flesh
    -0.66
    POSITIVE LOGITS
     amount
    0.64
     matrix
    0.59
     arous
    0.59
    weight
    0.57
     overtime
    0.57
     quarters
    0.56
     mov
    0.56
     department
    0.56
     spaceship
    0.55
     requirements
    0.54
    Activations Density 0.096%

    No Known Activations