© 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. 15840
    Prev
    Next
    INDEX
    Explanations

    information related to social and political protests

    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
    480
    +0.43
    2.2%
    266
    +0.10
    0.5%
    656
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.43
    -0.00
    59
    +0.10
    0.04
    266
    +0.09
    0.08
    Negative Logits
     dependency
    -0.78
    agate
    -0.75
     poisons
    -0.73
    conom
    -0.70
     Genetic
    -0.67
     solves
    -0.66
     depends
    -0.65
     Currently
    -0.64
    ulnerability
    -0.63
     relies
    -0.63
    POSITIVE LOGITS
     applause
    1.30
     attendees
    1.18
     podium
    1.13
     cheers
    1.12
     chanting
    1.10
     chants
    1.09
     spectators
    1.07
     microphones
    1.06
     cheering
    1.03
     loudspe
    1.02
    Activations Density 1.027%

    No Known Activations