© 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 12288
    9. 8-RES_FS12288-JB
    11. 1374
    Prev
    Next
    INDEX
    Explanations

    words related to technology features and products

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    371
    +0.10
    0.4%
    481
    +0.09
    0.4%
    436
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    24
    +0.10
    0.06
    659
    +0.09
    0.06
    232
    +0.09
    0.06
    Negative Logits
    ãĥ¯ãĥ³
    -0.86
    odi
    -0.83
    issues
    -0.80
    STD
    -0.80
    URA
    -0.79
    gdala
    -0.79
    ã
    -0.77
    INAL
    -0.77
    ept
    -0.76
    ¿½
    -0.76
    POSITIVE LOGITS
    ®
    0.91
     County
    0.86
     Chronicles
    0.85
     Manor
    0.84
     Cavern
    0.83
     Revis
    0.82
     Falls
    0.82
    hire
    0.82
    âĦ¢
    0.76
     Parish
    0.76
    Activations Density 0.625%

    No Known Activations