© 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. 16967
    Prev
    Next
    INDEX
    Explanations

    microwave and radiation measurements

    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
    67
    +0.10
    0.4%
    607
    +0.10
    0.4%
    352
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    67
    +0.10
    0.04
    252
    +0.10
    0.03
    238
    +0.09
    0.03
    Negative Logits
     backstage
    -0.69
     Digest
    -0.61
     Companion
    -0.60
     Sovere
    -0.60
     instructor
    -0.59
     spont
    -0.59
     itiner
    -0.59
     Orn
    -0.59
     Grateful
    -0.59
     advising
    -0.58
    POSITIVE LOGITS
    pees
    1.24
    abytes
    1.22
    illion
    1.21
    awatts
    1.15
    abyte
    1.09
    renheit
    1.06
    ousand
    1.04
    undred
    0.99
    rams
    0.95
    seconds
    0.94
    Activations Density 0.092%

    No Known Activations