© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 9679
    Prev
    Next
    INDEX
    Explanations

    lines of code from a programming language such as Rust, particularly focusing on methods related to parsing and tools used in testing

    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
    481
    +0.21
    1.0%
    584
    +0.11
    0.5%
    301
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.21
    0.05
    519
    +0.11
    0.04
    288
    +0.10
    0.06
    Negative Logits
    Ô
    -0.81
     Confederation
    -0.69
    æ©
    -0.67
    erest
    -0.62
     retaliate
    -0.62
    otropic
    -0.61
    anan
    -0.61
     tradem
    -0.60
    wikipedia
    -0.60
     curv
    -0.60
    POSITIVE LOGITS
    ummies
    0.69
    gery
    0.67
    rial
    0.66
    earances
    0.66
     Jury
    0.64
    rium
    0.64
    lete
    0.63
    manager
    0.62
     LR
    0.62
    undown
    0.61
    Activations Density 0.479%

    No Known Activations