© 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 12288
    9. 8-RES_FS12288-JB
    11. 10101
    Prev
    Next
    INDEX
    Explanations

    time references such as durations in minutes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    314
    +0.09
    0.4%
    359
    +0.09
    0.4%
    203
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    260
    +0.09
    0.06
    462
    +0.09
    0.05
    639
    +0.08
    0.05
    Negative Logits
     Balk
    -0.74
    worldly
    -0.70
    aida
    -0.69
    accompanied
    -0.64
     Kodi
    -0.63
     bait
    -0.63
    adesh
    -0.63
     subp
    -0.61
     prey
    -0.61
     scissors
    -0.61
    POSITIVE LOGITS
    degree
    1.00
    percent
    0.94
     MPH
    0.88
    ãĤ¼ãĤ¦ãĤ¹
    0.86
     percentile
    0.77
     Percent
    0.76
    ãĥīãĥ©ãĤ´ãĥ³
    0.75
    254
    0.75
    89
    0.74
    678
    0.74
    Activations Density 0.269%

    No Known Activations