© 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. 8724
    Prev
    Next
    INDEX
    Explanations

    links or references in the text

    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
    640
    +0.23
    1.1%
    679
    +0.13
    0.6%
    6
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.23
    0.07
    640
    +0.13
    0.06
    679
    +0.12
    0.06
    Negative Logits
     imperson
    -0.89
     awakening
    -0.87
     blowing
    -0.75
     forcing
    -0.75
     overpower
    -0.74
     acute
    -0.74
     recovering
    -0.73
     escaping
    -0.72
     plurality
    -0.72
     overpowered
    -0.72
    POSITIVE LOGITS
    <|endoftext|>
    1.41
     Alternatively
    1.38
     Also
    1.27
     Additionally
    1.16
     Otherwise
    1.15
     Downloads
    1.13
     Lastly
    1.08
    pdf
    1.07
     Includes
    1.05
     Originally
    1.05
    Activations Density 0.213%

    No Known Activations