© 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. 13782
    Prev
    Next
    INDEX
    Explanations

    stories related to conflict and violence

    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
    36
    +0.28
    1.4%
    393
    +0.13
    0.6%
    250
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.28
    0.19
    250
    +0.13
    0.09
    328
    +0.12
    0.07
    Negative Logits
     annually
    -0.75
    Own
    -0.71
    iments
    -0.70
    America
    -0.70
    NAS
    -0.68
     Effective
    -0.68
    Americans
    -0.68
     herein
    -0.66
     campuses
    -0.65
    SOURCE
    -0.65
    POSITIVE LOGITS
     handful
    1.13
     couple
    1.10
     few
    1.05
     nearby
    0.99
     flurry
    0.96
     bunch
    0.90
     brief
    0.88
     lot
    0.88
     friend
    0.87
     towel
    0.87
    Activations Density 0.434%

    No Known Activations