© 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 EvalsExportsAPI 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. 9108
    Prev
    Next
    INDEX
    Explanations

    website URLs and contact information

    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
    +0.15
    0.7%
    238
    +0.10
    0.5%
    291
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    435
    +0.15
    0.02
    145
    +0.10
    0.02
    462
    +0.09
    0.02
    Negative Logits
    =-=-=-=-
    -0.73
     afore
    -0.70
     TEXT
    -0.67
     inval
    -0.64
    ,,,,,,,,
    -0.64
     distance
    -0.62
     suffice
    -0.61
     اÙĦ
    -0.60
     anus
    -0.59
     commence
    -0.59
    POSITIVE LOGITS
    zx
    0.97
    qq
    0.95
    qa
    0.94
    q
    0.91
    nw
    0.89
    qi
    0.88
    zl
    0.86
    bh
    0.84
    yp
    0.82
    YY
    0.82
    Activations Density 0.020%

    No Known Activations