© 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. 16589
    Prev
    Next
    INDEX
    Explanations

    phrases related to politics and hacking

    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
    70
    +0.11
    0.5%
    684
    +0.10
    0.4%
    617
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.11
    0.06
    131
    +0.10
    0.04
    663
    +0.09
    0.03
    Negative Logits
    artifacts
    -0.77
    ifference
    -0.69
    paying
    -0.69
    ritional
    -0.67
    gression
    -0.65
    immune
    -0.65
    ographics
    -0.63
     imped
    -0.63
    commit
    -0.62
    warm
    -0.61
    POSITIVE LOGITS
     acronym
    1.55
     slang
    1.42
     coined
    1.41
     abbre
    1.36
     moniker
    1.35
     name
    1.32
     nickname
    1.29
     shorthand
    1.25
     term
    1.21
     initials
    1.19
    Activations Density 0.642%

    No Known Activations