© 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. 14960
    Prev
    Next
    INDEX
    Explanations

    the name "Mer" or "mer" which seems to appear in names and contexts related to military, politics, and conspiracy

    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
    373
    +0.11
    0.5%
    508
    +0.10
    0.5%
    763
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    763
    +0.11
    0.01
    284
    +0.10
    0.01
    556
    +0.10
    0.01
    Negative Logits
    ERO
    -0.74
    arial
    -0.63
    avorite
    -0.62
    numbered
    -0.60
    ors
    -0.60
     Strauss
    -0.59
    REG
    -0.58
    wn
    -0.58
    Benef
    -0.58
     Malfoy
    -0.57
    POSITIVE LOGITS
    gency
    1.24
    chant
    1.04
    maid
    1.02
    ikan
    0.94
    cia
    0.93
    gence
    0.93
    rill
    0.93
    zb
    0.91
    lund
    0.88
    mans
    0.88
    Activations Density 0.022%

    No Known Activations