© 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. 22618
    Prev
    Next
    INDEX
    Explanations

    negative assessments of ideas or options

    oai_token-act-pair · gpt-4o-miniTriggered 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
    512
    +0.13
    0.6%
    170
    +0.11
    0.5%
    234
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    512
    +0.13
    0.06
    170
    +0.11
    0.05
    718
    +0.10
    0.05
    Negative Logits
    ebted
    -0.81
     indebted
    -0.62
     curves
    -0.59
     envelop
    -0.59
     subcontract
    -0.59
     committees
    -0.59
     Lockheed
    -0.56
    picked
    -0.55
     Auditor
    -0.55
     Hew
    -0.55
    POSITIVE LOGITS
     preferable
    0.89
     heresy
    0.84
     advisable
    0.84
     prerequisite
    0.81
    $$$$
    0.81
     feasible
    0.78
     acceptable
    0.78
     icing
    0.76
    acea
    0.76
     blasphemy
    0.74
    Activations Density 0.283%

    No Known Activations