© 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. 9845
    Prev
    Next
    INDEX
    Explanations

    terms related to motor accidents or incidents

    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
    288
    +0.19
    0.9%
    27
    +0.13
    0.6%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    27
    +0.19
    0.02
    288
    +0.13
    0.02
    220
    +0.11
    0.02
    Negative Logits
     Foley
    -0.79
     Michaels
    -0.76
     Beg
    -0.74
     Carbuncle
    -0.72
     Merrill
    -0.71
     BET
    -0.69
     Ming
    -0.68
     Alibaba
    -0.67
     Antar
    -0.66
     Suc
    -0.66
    POSITIVE LOGITS
    cycle
    1.43
    bike
    1.35
    cycl
    1.20
    cycles
    1.17
    ized
    1.14
    otor
    1.06
    ised
    1.04
    wagen
    0.96
    puter
    0.95
    cars
    0.94
    Activations Density 0.054%

    No Known Activations