© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 15320
    Prev
    Next
    INDEX
    Explanations

    phrases related to movement and spatial relationships

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    218
    +0.13
    0.6%
    608
    +0.13
    0.6%
    442
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.13
    0.05
    123
    +0.13
    0.04
    218
    +0.09
    0.03
    Negative Logits
    gang
    -0.76
    zai
    -0.69
     timetable
    -0.66
     summons
    -0.64
    cpu
    -0.63
    innie
    -0.61
    yes
    -0.60
    ĺħ
    -0.58
    ector
    -0.58
    pick
    -0.58
    POSITIVE LOGITS
     adulthood
    1.13
     orbit
    0.91
     unch
    0.86
     infinity
    0.79
     adolescence
    0.79
     Valhalla
    0.77
     outer
    0.76
    ãĤ§
    0.75
    vernment
    0.73
     obscurity
    0.73
    Activations Density 0.082%

    No Known Activations