© 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 49152
    9. 8-RES_FS49152-JB
    11. 45531
    Prev
    Next
    INDEX
    Explanations

    code-related syntax elements, particularly those used in JavaScript-style coding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    756
    +0.12
    0.5%
    373
    +0.11
    0.5%
    153
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.12
    0.01
    656
    +0.11
    0.01
    244
    +0.10
    0.01
    Negative Logits
    ty
    -0.64
    gling
    -0.62
    lin
    -0.61
    CI
    -0.61
    ulate
    -0.60
    icka
    -0.60
    point
    -0.59
    compl
    -0.59
    illary
    -0.59
    rity
    -0.58
    POSITIVE LOGITS
    agna
    0.74
     "$:/
    0.73
     isEnabled
    0.69
    ãĥ¤
    0.69
    ãģ®ç
    0.68
    perse
    0.65
    %]
    0.63
    onent
    0.62
    iddler
    0.61
    à¦
    0.61
    Activations Density 0.011%

    No Known Activations