© 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. 1629
    Prev
    Next
    INDEX
    Explanations

    superlatives or comparative adjectives like "biggest," "greatest," or "oldest"

    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
    122
    +0.12
    0.5%
    116
    +0.11
    0.5%
    446
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    122
    +0.12
    0.04
    185
    +0.11
    0.04
    616
    +0.10
    0.04
    Negative Logits
    rons
    -0.75
    cs
    -0.74
     Cox
    -0.72
    pload
    -0.72
    akov
    -0.70
    ttes
    -0.69
    kins
    -0.69
    rox
    -0.68
    fen
    -0.68
    kin
    -0.68
    POSITIVE LOGITS
     superpower
    0.91
     tourist
    0.79
     industrial
    0.76
     agricultural
    0.73
     export
    0.72
     spy
    0.72
     foray
    0.71
     premier
    0.68
     commodity
    0.67
     culinary
    0.67
    Activations Density 0.099%

    No Known Activations