© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9351
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to specific individuals, locations, and organizations

    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
    216
    +0.13
    0.6%
    19
    +0.13
    0.6%
    128
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    128
    +0.13
    0.03
    19
    +0.13
    0.03
    22
    +0.12
    0.02
    Negative Logits
     thumbs
    -0.68
    hound
    -0.67
    staking
    -0.63
     deaf
    -0.63
     Peel
    -0.62
     consultation
    -0.62
     calming
    -0.62
     prepar
    -0.61
     competent
    -0.61
    perty
    -0.60
    POSITIVE LOGITS
    retch
    1.14
    oppable
    1.12
    oppers
    1.10
    rict
    1.09
    amped
    1.05
    alker
    1.04
    itute
    1.03
    upid
    1.01
    hetics
    1.01
    itution
    1.00
    Activations Density 0.030%

    No Known Activations