© 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. 7492
    Prev
    Next
    INDEX
    Explanations

    phrases related to technical details or instructions, specifically focusing on measurements, directions, and specifications

    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
    31
    +0.09
    0.4%
    747
    +0.09
    0.4%
    388
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    747
    +0.09
    0.01
    126
    +0.09
    0.01
    619
    +0.09
    0.02
    Negative Logits
     tremend
    -0.72
    aminer
    -0.72
     mathemat
    -0.66
     estab
    -0.64
     intrins
    -0.63
     kitten
    -0.62
    hered
    -0.60
     scant
    -0.60
    ©¶æ
    -0.57
     millisec
    -0.57
    POSITIVE LOGITS
    enged
    1.20
    owing
    1.19
    iance
    1.16
    adium
    1.16
    enge
    1.16
    ocated
    1.14
    ength
    1.13
    ergic
    1.13
    ocation
    1.08
    enges
    1.07
    Activations Density 0.021%

    No Known Activations