© 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 24576
    9. 8-RES_FS24576-JB
    11. 18902
    Prev
    Next
    INDEX
    Explanations

    timestamp formats

    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
    479
    +0.15
    0.7%
    308
    +0.14
    0.6%
    145
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.15
    0.06
    479
    +0.14
    0.03
    308
    +0.13
    0.03
    Negative Logits
    rac
    -0.80
    camp
    -0.78
    arg
    -0.77
    college
    -0.74
     marqu
    -0.73
     diving
    -0.72
    opian
    -0.72
     laser
    -0.72
     migr
    -0.71
    cooked
    -0.71
    POSITIVE LOGITS
     Less
    1.31
     Same
    1.23
     Only
    1.23
     Remove
    1.23
     Absent
    1.22
     Given
    1.22
     Enough
    1.22
     Not
    1.22
     Percent
    1.20
     Nope
    1.20
    Activations Density 0.234%

    No Known Activations