© 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 24576
    9. 8-RES_FS24576-JB
    11. 11909
    Prev
    Next
    INDEX
    Explanations

    details about future plans, updates, and policies

    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
    533
    +0.10
    0.5%
    254
    +0.10
    0.5%
    679
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    519
    +0.10
    0.03
    756
    +0.10
    -0.01
    714
    +0.10
    0.02
    Negative Logits
     Samar
    -0.73
     Patreon
    -0.65
     marg
    -0.64
     charity
    -0.63
     Meter
    -0.60
     meter
    -0.60
     Charity
    -0.59
     Utility
    -0.59
     gif
    -0.58
     nutshell
    -0.58
    POSITIVE LOGITS
     inher
    0.88
    20439
    0.79
    morrow
    0.79
    hran
    0.72
    soon
    0.72
    hop
    0.72
    ichever
    0.70
    etheus
    0.70
    anticipated
    0.70
     inherit
    0.69
    Activations Density 0.375%

    No Known Activations