© 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. 1875
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Ta" with varying levels of activation, potentially referring to a specific individual or entity named Ta

    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
    57
    +0.11
    0.5%
    718
    +0.10
    0.5%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.02
    57
    +0.10
    0.01
    682
    +0.09
    0.01
    Negative Logits
    lessly
    -0.78
     OCD
    -0.65
     atmosphere
    -0.64
     reception
    -0.63
     shortages
    -0.62
     optics
    -0.60
     injection
    -0.59
     CTR
    -0.59
     Governors
    -0.59
     ANGEL
    -0.59
    POSITIVE LOGITS
    plin
    1.32
    iba
    1.26
    vern
    1.14
    pling
    1.08
    fts
    1.05
    oshi
    0.99
    uren
    0.98
    onga
    0.98
    ifa
    0.96
    uman
    0.96
    Activations Density 0.026%

    No Known Activations