© 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. 1256
    Prev
    Next
    INDEX
    Explanations

    locations or place names.

    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
    99
    +0.16
    0.7%
    211
    +0.15
    0.7%
    267
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    211
    +0.16
    0.01
    99
    +0.15
    0.02
    219
    +0.11
    0.02
    Negative Logits
    efully
    -0.83
    manship
    -0.74
    ylan
    -0.71
    lyak
    -0.70
    oric
    -0.69
    utan
    -0.67
     breaker
    -0.64
    ighth
    -0.64
    orically
    -0.64
    Wan
    -0.64
    POSITIVE LOGITS
     Pike
    0.87
    ville
    0.81
     Speedway
    0.80
     Heights
    0.79
    sburgh
    0.78
     Parish
    0.78
     Slug
    0.78
     Ferry
    0.77
     Horror
    0.76
     Courier
    0.74
    Activations Density 0.026%

    No Known Activations