© 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. 21978
    Prev
    Next
    INDEX
    Explanations

    references to sports teams and rankings

    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
    167
    +0.12
    0.6%
    26
    +0.12
    0.5%
    754
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    26
    +0.12
    0.01
    742
    +0.12
    0.01
    754
    +0.11
    0.01
    Negative Logits
    forcing
    -0.74
    ricular
    -0.70
    staking
    -0.68
    fully
    -0.68
    ful
    -0.66
     Gamble
    -0.66
    liest
    -0.66
    nces
    -0.64
    fulness
    -0.62
    lished
    -0.61
    POSITIVE LOGITS
    DN
    0.93
    NP
    0.90
    ML
    0.89
    TON
    0.87
    TP
    0.84
    MX
    0.83
    NL
    0.81
    endon
    0.81
    RS
    0.81
    ounge
    0.80
    Activations Density 0.015%

    No Known Activations