Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 19586
    Prev
    Next
    INDEX
    Explanations

    sections related to political statements and international events

    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
    390
    +0.09
    0.4%
    572
    +0.09
    0.4%
    66
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    541
    +0.09
    0.03
    390
    +0.09
    0.02
    435
    +0.09
    0.02
    Negative Logits
    ridge
    -0.77
    ites
    -0.71
    azon
    -0.70
    airo
    -0.69
    lang
    -0.68
    itters
    -0.68
    orrow
    -0.66
    anwhile
    -0.64
    ãĢı
    -0.64
    anta
    -0.63
    POSITIVE LOGITS
     insur
    0.95
     punitive
    0.79
     Byzantine
    0.76
     bipartisan
    0.74
     rudimentary
    0.74
     concerted
    0.73
     systemic
    0.72
     irreversible
    0.72
     erroneous
    0.71
     miraculous
    0.71
    Activations Density 0.087%

    No Known Activations