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

    Neuronpedia

    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. 19453
    Prev
    Next
    INDEX
    Explanations

    dialogue interactions between two characters, specifically focusing on the back-and-forth exchanges

    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
    145
    +0.13
    0.6%
    432
    +0.10
    0.5%
    496
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.04
    145
    +0.10
    0.03
    472
    +0.10
    0.02
    Negative Logits
    escription
    -0.87
    ascus
    -0.81
    eatures
    -0.78
    inement
    -0.72
    æ©
    -0.71
    avorite
    -0.71
    abad
    -0.70
    irements
    -0.69
    lection
    -0.67
    velop
    -0.66
    POSITIVE LOGITS
     Yeah
    1.47
     Alright
    1.43
     Hmm
    1.31
    Yeah
    1.26
     Exactly
    1.25
     Okay
    1.23
     Oh
    1.17
     Nope
    1.12
     Huh
    1.12
     Yep
    1.10
    Activations Density 0.100%

    No Known Activations