© 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. 6159
    Prev
    Next
    INDEX
    Explanations

    words related to natural disasters, specifically tsunamis

    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
    267
    +0.12
    0.6%
    373
    +0.11
    0.5%
    282
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    406
    +0.11
    0.03
    476
    +0.11
    0.02
    Negative Logits
     Madden
    -0.71
    ãĤ¼
    -0.66
    Ö¼
    -0.66
     abiding
    -0.65
    INC
    -0.63
     accur
    -0.61
    ortment
    -0.59
     Peb
    -0.58
     refres
    -0.58
     Magicka
    -0.58
    POSITIVE LOGITS
    geon
    1.35
    geons
    1.24
    nel
    1.23
    culosis
    1.18
    culus
    1.14
    ned
    1.12
    nels
    1.08
    ners
    1.07
    iversity
    1.06
    iverse
    1.04
    Activations Density 0.030%

    No Known Activations