© 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. 8511
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology, specifically focusing on hardware components and technical specifications

    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
    565
    +0.11
    0.5%
    654
    +0.10
    0.5%
    460
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    512
    +0.11
    0.06
    718
    +0.10
    0.06
    102
    +0.10
    0.07
    Negative Logits
     Secrets
    -0.69
     Saud
    -0.69
     Lies
    -0.68
     Visitors
    -0.68
     Brow
    -0.67
     Ow
    -0.66
     Lists
    -0.66
     Intellectual
    -0.64
     Roger
    -0.63
    igious
    -0.63
    POSITIVE LOGITS
     preferable
    1.58
     feasible
    1.31
     viable
    1.23
     preferred
    1.23
     advisable
    1.22
     recommended
    1.09
     safer
    1.09
     impractical
    1.09
     favoured
    1.08
     desirable
    1.06
    Activations Density 0.637%

    No Known Activations