© 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 3072
    5. 8-RES_FS3072-JB
    6. 50
    Prev
    Next
    INDEX
    Explanations

    keywords related to programming languages and frameworks, specifically focusing on Haskell and web development

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    447
    +0.11
    0.5%
    57
    +0.11
    0.5%
    480
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    22
    +0.11
    0.10
    362
    +0.11
    0.11
    616
    +0.11
    0.08
    Negative Logits
     accordingly
    -0.92
    jri
    -0.69
    ONEY
    -0.68
     Vaugh
    -0.65
    SpaceEngineers
    -0.65
    efully
    -0.65
    omething
    -0.64
    çİĭ
    -0.64
    WHAT
    -0.64
     scrut
    -0.63
    POSITIVE LOGITS
     nutshell
    1.16
     Hunts
    0.76
     Depth
    0.75
    ked
    0.75
    verts
    0.73
     Atmosp
    0.73
     Movies
    0.73
     Context
    0.71
     Pigs
    0.71
     Action
    0.70
    Activations Density 1.816%

    No Known Activations