© 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 1536
    9. 8-RES_FS1536-JB
    11. 44
    Prev
    Next
    INDEX
    Explanations

    phrases related to controversial topics or polarizing opinions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    391
    +0.13
    0.6%
    373
    +0.13
    0.6%
    306
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    267
    +0.13
    0.10
    389
    +0.13
    0.08
    306
    +0.11
    0.09
    Negative Logits
    inction
    -0.77
    ourse
    -0.76
    icipated
    -0.74
    erness
    -0.73
    İĭ
    -0.71
    arcity
    -0.70
    actionDate
    -0.69
    è¦ļéĨĴ
    -0.67
    ãĥį
    -0.67
    result
    -0.65
    POSITIVE LOGITS
     yeah
    1.22
    yeah
    1.11
     kidding
    1.11
     fuck
    1.03
     huh
    0.97
     Yeah
    0.96
     Fuck
    0.94
    oooo
    0.94
     crap
    0.94
     dude
    0.92
    Activations Density 2.649%

    No Known Activations