© 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 12288
    5. 8-RES_FS12288-JB
    6. 9530
    Prev
    Next
    INDEX
    Explanations

    political terms and concepts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    164
    +0.10
    0.5%
    393
    +0.10
    0.5%
    269
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.06
    476
    +0.10
    0.03
    393
    +0.10
    0.05
    Negative Logits
     spoilers
    -0.73
    obyl
    -0.71
    sensitive
    -0.71
    strike
    -0.69
    pmwiki
    -0.69
     lawfully
    -0.68
     agre
    -0.68
    TPPStreamerBot
    -0.67
    answered
    -0.67
    hazard
    -0.66
    POSITIVE LOGITS
     Northeast
    0.90
     Europe
    0.88
     antiquity
    0.86
     1930
    0.85
     adolescence
    0.84
     Appalach
    0.84
     Eastern
    0.83
     southwestern
    0.82
     infancy
    0.80
     Southeast
    0.79
    Activations Density 0.282%

    No Known Activations