© 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 3072
    9. 8-RES_FS3072-JB
    11. 246
    Prev
    Next
    INDEX
    Explanations

    financial terms and names, possibly related to stock markets and companies

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    337
    +0.14
    0.6%
    525
    +0.11
    0.5%
    406
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    406
    +0.14
    0.09
    232
    +0.11
    0.08
    373
    +0.10
    -0.05
    Negative Logits
    ħĭ
    -0.85
    INAL
    -0.83
    querque
    -0.78
    actionDate
    -0.77
    vantage
    -0.76
    ĺħ
    -0.76
    £ı
    -0.73
    ãĥ¼ãĥĨ
    -0.70
     srfAttach
    -0.70
    ļé
    -0.69
    POSITIVE LOGITS
    nesday
    0.96
    ipeg
    0.87
    ÅĤ
    0.81
    lich
    0.81
    lde
    0.79
    wright
    0.78
    owers
    0.73
    reens
    0.73
    hester
    0.72
    itzer
    0.71
    Activations Density 1.393%

    No Known Activations