© 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. 1650
    Prev
    Next
    INDEX
    Explanations

    organizations or groups of people and places

    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
    481
    +0.15
    0.7%
    416
    +0.10
    0.5%
    55
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    197
    +0.15
    0.03
    318
    +0.10
    0.03
    506
    +0.09
    0.03
    Negative Logits
    ogle
    -0.73
    nce
    -0.63
    pta
    -0.62
    arah
    -0.62
    =-=-
    -0.61
     Newsletter
    -0.61
    trial
    -0.60
     quarry
    -0.60
    utory
    -0.59
     Psal
    -0.58
    POSITIVE LOGITS
     sexes
    1.13
     halves
    1.06
     equally
    0.99
     simultaneously
    0.97
     sides
    0.97
     mutually
    0.85
     concurrently
    0.84
     individually
    0.82
     alike
    0.82
    ieth
    0.80
    Activations Density 0.156%

    No Known Activations