© 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. 4279
    Prev
    Next
    INDEX
    Explanations

    Chinese characters or symbols

    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
    +0.12
    0.6%
    228
    +0.11
    0.5%
    284
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.12
    0.02
    406
    +0.11
    0.01
    103
    +0.11
    0.01
    Negative Logits
    pole
    -0.60
     equival
    -0.58
     adoption
    -0.56
     convergence
    -0.54
    eele
    -0.54
     bonding
    -0.54
     awareness
    -0.53
     traff
    -0.53
    mand
    -0.53
    uese
    -0.53
    POSITIVE LOGITS
    Ķ
    1.13
    ľ
    1.11
    ļ
    1.08
    ī
    1.07
    º
    1.06
    ĺ
    1.06
    ¼
    1.05
    ħ
    1.04
    ¿
    1.04
    IJ
    1.03
    Activations Density 0.021%

    No Known Activations