© 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 24576
    9. 8-RES_FS24576-JB
    11. 19641
    Prev
    Next
    INDEX
    Explanations

    words related to instructions or steps required to complete a task

    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
    408
    +0.14
    0.6%
    362
    +0.10
    0.5%
    497
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.14
    0.03
    347
    +0.10
    0.02
    511
    +0.09
    0.02
    Negative Logits
    hey
    -0.93
    arer
    -0.85
    peak
    -0.82
    ĸļ
    -0.79
    gemony
    -0.79
    Cola
    -0.78
    lan
    -0.78
    arthed
    -0.78
    oche
    -0.77
    ruary
    -0.76
    POSITIVE LOGITS
     amount
    1.16
     paperwork
    1.15
     ingredients
    1.11
     components
    1.04
     permissions
    1.04
     materials
    0.99
     amounts
    0.98
     portions
    0.96
     portion
    0.94
     parts
    0.93
    Activations Density 0.084%

    No Known Activations