© 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. 17134
    Prev
    Next
    INDEX
    Explanations

    phrases related to making logical or rational sense

    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
    410
    +0.11
    0.5%
    638
    +0.09
    0.4%
    517
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.11
    0.03
    410
    +0.09
    0.02
    638
    +0.08
    0.02
    Negative Logits
    ħĭ
    -0.69
     diligently
    -0.69
     mathemat
    -0.63
     manually
    -0.62
    anwhile
    -0.61
     aggressively
    -0.61
    IDS
    -0.60
    rompt
    -0.60
    Ħ¢
    -0.59
     directions
    -0.59
    POSITIVE LOGITS
     difference
    0.98
     sense
    0.93
    sense
    0.92
     Difference
    0.88
     mockery
    0.86
     impression
    0.82
     noise
    0.74
     dent
    0.72
     Sense
    0.71
     wearer
    0.70
    Activations Density 0.096%

    No Known Activations