© 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 1536
    9. 8-RES_FS1536-JB
    11. 1072
    Prev
    Next
    INDEX
    Explanations

    phrases related to responsibilities and involvements in actions or decisions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    309
    +0.10
    0.5%
    259
    +0.10
    0.4%
    620
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.10
    0.15
    362
    +0.10
    0.12
    250
    +0.10
    0.06
    Negative Logits
    types
    -0.76
    ibl
    -0.72
    onyms
    -0.70
    nir
    -0.69
    erm
    -0.68
     acquaintances
    -0.65
     horm
    -0.64
    raph
    -0.63
    rentices
    -0.63
    Fish
    -0.62
    POSITIVE LOGITS
     shaping
    1.08
     overseeing
    1.06
     orchestr
    0.89
     drafting
    0.86
     facilitating
    0.86
     organising
    0.85
     approving
    0.84
     determining
    0.84
     steering
    0.84
     brunt
    0.82
    Activations Density 3.338%

    No Known Activations