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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 7336
    Prev
    Next
    INDEX
    Explanations

    statements related to conflicts and agreements

    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
    314
    +0.16
    0.9%
    86
    +0.09
    0.5%
    680
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    13
    +0.16
    0.06
    756
    +0.09
    -0.05
    86
    +0.08
    0.04
    Negative Logits
     Pinball
    -0.64
     brill
    -0.62
     sprite
    -0.61
     gorgeous
    -0.61
     weird
    -0.60
     Wander
    -0.59
     Slayer
    -0.58
     Firefly
    -0.58
     lovely
    -0.57
     coolest
    -0.57
    POSITIVE LOGITS
    partisan
    0.74
    20439
    0.72
     jurisdiction
    0.71
    Government
    0.70
     redacted
    0.68
    displayText
    0.67
    Gov
    0.67
    ashington
    0.66
    Federal
    0.66
    judicial
    0.66
    Activations Density 1.860%

    No Known Activations