© 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. 9817
    Prev
    Next
    INDEX
    Explanations

    terms related to time sequences

    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
    218
    +0.10
    0.4%
    157
    +0.10
    0.4%
    662
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    157
    +0.10
    0.04
    662
    +0.10
    0.03
    147
    +0.09
    0.03
    Negative Logits
    pta
    -0.68
    adiator
    -0.67
    natureconservancy
    -0.67
    ©¶æ¥µ
    -0.64
    Si
    -0.63
    dozen
    -0.61
    atron
    -0.60
    vantage
    -0.59
    immer
    -0.59
     lash
    -0.58
    POSITIVE LOGITS
    rely
    1.82
     inception
    0.87
     1945
    0.81
     1979
    0.77
     1999
    0.77
     1995
    0.77
     2006
    0.76
     January
    0.76
     2005
    0.75
     2001
    0.75
    Activations Density 0.051%

    No Known Activations