© 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 49152
    9. 8-RES_FS49152-JB
    11. 37325
    Prev
    Next
    INDEX
    Explanations

    phrases and words associated with emotional expressions and moments of awareness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    56
    +0.10
    0.5%
    58
    +0.10
    0.4%
    284
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    40
    +0.10
    0.03
    132
    +0.10
    0.02
    56
    +0.10
    0.03
    Negative Logits
    orthy
    -0.84
    Published
    -0.79
    Guest
    -0.69
    ourses
    -0.65
    atre
    -0.64
    dule
    -0.64
    uthor
    -0.64
    missions
    -0.63
    ufact
    -0.62
    Recommend
    -0.62
    POSITIVE LOGITS
     stare
    0.81
     pesky
    0.80
     hump
    0.78
     thing
    0.78
     alone
    0.77
    milo
    0.72
     smile
    0.71
     itch
    0.70
     popping
    0.67
     grin
    0.67
    Activations Density 0.162%

    No Known Activations