© 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 12288
    9. 8-RES_FS12288-JB
    11. 10737
    Prev
    Next
    INDEX
    Explanations

    personal pronouns related to addressing or referring to the reader

    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
    49
    +0.12
    0.5%
    736
    +0.11
    0.5%
    167
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    49
    +0.12
    0.09
    736
    +0.11
    0.07
    354
    +0.11
    0.07
    Negative Logits
    ¿½
    -0.70
    shore
    -0.68
     Gamb
    -0.63
     airs
    -0.62
     srfAttach
    -0.60
    ãĥ³ãĤ¸
    -0.59
    wrapper
    -0.58
    assemb
    -0.57
     stemming
    -0.56
    ipal
    -0.56
    POSITIVE LOGITS
    're
    1.45
    'll
    1.32
    've
    1.25
     guessed
    1.19
     guys
    1.12
    'd
    1.09
    tub
    1.02
    imar
    0.97
     know
    0.96
     mileage
    0.95
    Activations Density 0.188%

    No Known Activations