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

    Neuronpedia

    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 12288
    9. 8-RES_FS12288-JB
    11. 10807
    Prev
    Next
    INDEX
    Explanations

    concepts or explanations involving averages or means in a mathematical context

    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
    261
    +0.11
    0.5%
    296
    +0.10
    0.5%
    489
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.11
    0.05
    582
    +0.10
    0.03
    105
    +0.10
    0.04
    Negative Logits
    inventoryQuantity
    -1.05
     redes
    -0.77
    "},
    -0.76
    notations
    -0.73
    vre
    -0.73
     understatement
    -0.71
    ="#
    -0.70
    ãĤ¦ãĤ¹
    -0.69
     merits
    -0.69
    ;;;;;;;;;;;;
    -0.69
    POSITIVE LOGITS
     Suddenly
    0.89
     Suppose
    0.88
     suddenly
    0.86
    Suddenly
    0.81
     approached
    0.79
     randomly
    0.77
     Plaintiff
    0.77
    uddenly
    0.76
     somebody
    0.75
     casually
    0.74
    Activations Density 1.052%

    No Known Activations