© 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. 22
    Prev
    Next
    INDEX
    Explanations

    legal and political terms or phrases

    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
    314
    +0.26
    1.3%
    640
    +0.22
    1.1%
    160
    +0.19
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.26
    0.15
    314
    +0.22
    0.04
    627
    +0.19
    0.09
    Negative Logits
     retard
    -0.77
     lull
    -0.76
     clin
    -0.73
     intensive
    -0.72
     daily
    -0.72
     wildlife
    -0.71
     inund
    -0.71
     everyday
    -0.70
     wre
    -0.70
     glare
    -0.69
    POSITIVE LOGITS
    sic
    1.66
    including
    1.59
    see
    1.58
    excluding
    1.55
    formerly
    1.55
    which
    1.55
    although
    1.51
    except
    1.50
    also
    1.47
    aka
    1.47
    Activations Density 0.482%

    No Known Activations