© 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. 18631
    Prev
    Next
    INDEX
    Explanations

    descriptions related to construction or completion stages of projects

    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
    207
    +0.12
    0.5%
    338
    +0.11
    0.5%
    455
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    207
    +0.12
    0.03
    333
    +0.11
    0.03
    339
    +0.10
    0.03
    Negative Logits
    aca
    -0.66
     instantly
    -0.63
     Lawn
    -0.62
    facts
    -0.62
    flat
    -0.60
    soc
    -0.60
     Kop
    -0.59
     Hebdo
    -0.58
     Mund
    -0.58
     Polic
    -0.57
    POSITIVE LOGITS
     preparations
    0.97
     infancy
    0.95
     stages
    0.88
     rebuilding
    0.83
     phases
    0.83
     reorgan
    0.82
     puberty
    0.82
     phase
    0.82
     decom
    0.82
     Phase
    0.81
    Activations Density 0.182%

    No Known Activations