© 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. 7
    Prev
    Next
    INDEX
    Explanations

    mentions of costs, expenses, or financial aspects

    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
    285
    +0.12
    0.5%
    612
    +0.11
    0.5%
    509
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    285
    +0.12
    0.06
    35
    +0.11
    0.07
    439
    +0.11
    0.07
    Negative Logits
    vironment
    -0.82
    ullivan
    -0.74
     Klu
    -0.70
    unin
    -0.69
     Beard
    -0.69
    itialized
    -0.67
    jun
    -0.67
    riter
    -0.66
    aug
    -0.65
    urst
    -0.65
    POSITIVE LOGITS
    umed
    1.30
     incurred
    1.14
     prohib
    1.13
    umers
    1.10
    liest
    1.07
    uming
    1.05
    umer
    1.00
    lier
    0.96
     overrun
    0.96
    payer
    0.94
    Activations Density 1.372%

    No Known Activations