© 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. 1468
    Prev
    Next
    INDEX
    Explanations

    instances of the word "toss" with varying levels of intensity in the text

    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
    251
    +0.10
    0.5%
    101
    +0.09
    0.4%
    193
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.10
    0.01
    251
    +0.09
    0.01
    469
    +0.09
    0.01
    Negative Logits
    KT
    -0.68
    HCR
    -0.64
    xia
    -0.61
    éĸ
    -0.61
    otype
    -0.60
     firsthand
    -0.58
    MAL
    -0.58
    PRES
    -0.58
     DEBUG
    -0.58
     livest
    -0.58
    POSITIVE LOGITS
     toss
    1.00
     aside
    0.97
     overboard
    0.96
    weed
    0.87
     tossing
    0.82
     tossed
    0.81
    bowl
    0.80
     dab
    0.76
    stakes
    0.76
    enger
    0.74
    Activations Density 0.034%

    No Known Activations