© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 4343
    Prev
    Next
    INDEX
    Explanations

    names or titles abbreviated with a letter followed by a number

    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
    373
    +0.18
    0.8%
    631
    +0.09
    0.4%
    45
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.18
    -0.01
    697
    +0.09
    0.01
    701
    +0.09
    0.01
    Negative Logits
    00007
    -0.62
     Hurricanes
    -0.59
     Scion
    -0.58
     Outlook
    -0.58
     debit
    -0.57
    interrupted
    -0.56
    jri
    -0.56
    killer
    -0.56
     Chiefs
    -0.56
     blazing
    -0.54
    POSITIVE LOGITS
    urnal
    0.77
    alon
    0.74
    ilus
    0.69
    ãĤ¿
    0.68
    ophy
    0.68
    bidden
    0.67
    aly
    0.67
    auri
    0.66
    rical
    0.66
    otally
    0.66
    Activations Density 0.032%

    No Known Activations