© 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. 335
    Prev
    Next
    INDEX
    Explanations

    examples of the word "Anyone"

    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
    +0.13
    0.6%
    497
    +0.11
    0.5%
    492
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.13
    0.03
    735
    +0.11
    0.03
    492
    +0.10
    0.02
    Negative Logits
    iger
    -0.66
     Hound
    -0.65
    tnc
    -0.65
    bows
    -0.65
    irth
    -0.65
    ocamp
    -0.64
    ories
    -0.63
    rament
    -0.63
    urations
    -0.62
    ulk
    -0.61
    POSITIVE LOGITS
     else
    1.79
    else
    1.35
     Else
    1.28
    Else
    1.26
     who
    1.03
    who
    0.96
     wishing
    0.95
     subscribed
    0.91
    soever
    0.91
    THING
    0.89
    Activations Density 0.044%

    No Known Activations