© 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 1536
    9. 8-RES_FS1536-JB
    11. 43
    Prev
    Next
    INDEX
    Explanations

    quotes spoken or recounted by someone

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    756
    +0.10
    0.5%
    495
    +0.10
    0.4%
    180
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    180
    +0.10
    0.10
    468
    +0.10
    0.09
    766
    +0.10
    0.08
    Negative Logits
    ãĥİ
    -0.75
    paralle
    -0.67
    atible
    -0.66
    otin
    -0.65
     conflic
    -0.61
     ILCS
    -0.61
    fet
    -0.61
    thur
    -0.59
    Fit
    -0.57
    Appearances
    -0.57
    POSITIVE LOGITS
     sarcast
    1.09
     rhet
    1.00
     bluntly
    0.97
     diplom
    0.90
     emphatically
    0.85
     referring
    0.80
     adding
    0.78
     incred
    0.76
     chuck
    0.74
     confidently
    0.74
    Activations Density 0.964%

    No Known Activations