© 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 24576
    9. 8-RES_FS24576-JB
    11. 19782
    Prev
    Next
    INDEX
    Explanations

    emotional or opinionated expressions related to personal experiences and interactions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    373
    +0.13
    0.6%
    370
    +0.11
    0.5%
    622
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    297
    +0.13
    0.06
    720
    +0.11
    0.07
    43
    +0.10
    0.06
    Negative Logits
    utive
    -0.63
     Finder
    -0.61
    rons
    -0.59
    utral
    -0.59
    rets
    -0.59
    oret
    -0.59
    isbury
    -0.58
    âĢİ
    -0.58
     Mald
    -0.57
     Falk
    -0.57
    POSITIVE LOGITS
    but
    0.93
     namely
    0.90
     viz
    0.90
     but
    0.88
     haha
    0.88
    unless
    0.87
     albeit
    0.87
    BUT
    0.86
     BUT
    0.85
     Anyway
    0.81
    Activations Density 0.426%

    No Known Activations