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

    Neuronpedia

    Jacobian LensNEW
    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. 9001
    Prev
    Next
    INDEX
    Explanations

    words related to social issues such as displacement, poverty, and crisis

    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
    671
    +0.10
    0.5%
    131
    +0.10
    0.5%
    292
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    190
    +0.10
    0.05
    616
    +0.10
    0.04
    284
    +0.09
    0.04
    Negative Logits
     magazines
    -0.81
     dishes
    -0.78
     weddings
    -0.76
    ateurs
    -0.75
    ovies
    -0.73
     toilets
    -0.72
     trailers
    -0.72
     thugs
    -0.72
     idiots
    -0.71
     mates
    -0.70
    POSITIVE LOGITS
    yssey
    0.83
     equation
    0.76
    urrence
    0.76
    itudinal
    0.74
    gap
    0.74
     timetable
    0.73
    road
    0.73
     epidemic
    0.71
    auntlet
    0.71
    problem
    0.70
    Activations Density 0.338%

    No Known Activations