© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10236
    Prev
    Next
    INDEX
    Explanations

    links to instructional resources or tutorials

    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
    +0.11
    0.5%
    505
    +0.10
    0.4%
    267
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.11
    0.05
    23
    +0.10
    0.04
    267
    +0.09
    0.04
    Negative Logits
     exorc
    -0.70
    leased
    -0.70
     reservation
    -0.66
     troubles
    -0.65
     annihil
    -0.65
     tremend
    -0.64
    ecided
    -0.64
    isman
    -0.63
    alog
    -0.62
     insurer
    -0.62
    POSITIVE LOGITS
     https
    1.32
     http
    1.28
    http
    1.03
    https
    1.02
    ↵Âł
    1.02
     Logged
    0.98
    ↵↵
    0.92
     Provided
    0.91
     âĹı
    0.90
    <|endoftext|>
    0.89
    Activations Density 0.121%

    No Known Activations