© 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. 24455
    Prev
    Next
    INDEX
    Explanations

    phrases expressing negative sentiments or criticisms

    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
    480
    +0.17
    0.8%
    314
    +0.13
    0.6%
    327
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    675
    +0.17
    0.12
    584
    +0.13
    0.10
    720
    +0.09
    0.10
    Negative Logits
     myster
    -0.69
     Folk
    -0.69
     soph
    -0.68
     elig
    -0.63
     scholars
    -0.63
     filler
    -0.63
     astronomers
    -0.62
     princ
    -0.62
     trainers
    -0.61
     leaps
    -0.58
    POSITIVE LOGITS
    ï¸ı
    0.94
    efe
    0.88
    ihad
    0.81
    sic
    0.79
    vernment
    0.79
    should
    0.78
    selves
    0.78
    resent
    0.78
    âĢİ
    0.76
    killed
    0.76
    Activations Density 0.425%

    No Known Activations