© 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. 9834
    Prev
    Next
    INDEX
    Explanations

    terms related to computer caching systems

    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
    447
    +0.15
    0.7%
    450
    +0.10
    0.4%
    630
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    461
    +0.15
    0.03
    288
    +0.10
    0.05
    231
    +0.09
    0.04
    Negative Logits
    Balt
    -0.74
     Carlton
    -0.69
     NCT
    -0.66
     Balt
    -0.65
     thw
    -0.64
     Canterbury
    -0.64
     Hayden
    -0.64
     Went
    -0.60
     Norton
    -0.59
    endale
    -0.58
    POSITIVE LOGITS
     involves
    1.11
     theorists
    1.06
     differs
    1.05
     consists
    1.05
     refers
    1.04
     implies
    1.04
     entails
    1.00
     tends
    0.98
     occurs
    0.94
     means
    0.92
    Activations Density 0.264%

    No Known Activations