© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9885
    Prev
    Next
    INDEX
    Explanations

    discussions about hypothetical scenarios and philosophical reflections

    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
    87
    +0.29
    1.4%
    756
    +0.14
    0.7%
    24
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.29
    -0.04
    756
    +0.14
    -0.02
    283
    +0.11
    0.02
    Negative Logits
    ebin
    -0.73
     WATCH
    -0.70
    Update
    -0.64
     HERE
    -0.61
     Adds
    -0.61
    osponsors
    -0.60
     Yesterday
    -0.60
    obook
    -0.59
    ctuary
    -0.58
     UPDATE
    -0.58
    POSITIVE LOGITS
     tended
    0.98
     mattered
    0.98
     depended
    0.93
     lacked
    0.91
     resembled
    0.83
     consisted
    0.82
     smelled
    0.81
     cared
    0.81
     knew
    0.81
     seemed
    0.81
    Activations Density 2.334%

    No Known Activations