© 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 768
    9. 8-RES_FS768-JB
    11. 644
    Prev
    Next
    INDEX
    Explanations

    step-by-step instructions or procedures

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    640
    +0.37
    1.8%
    615
    +0.10
    0.5%
    314
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.37
    0.32
    143
    +0.10
    0.16
    615
    +0.10
    0.15
    Negative Logits
    hement
    -0.88
     Cent
    -0.69
     aging
    -0.63
     ageing
    -0.62
     honoured
    -0.62
     manoeuv
    -0.60
     hiber
    -0.59
     scrap
    -0.59
     estranged
    -0.58
    ensibly
    -0.57
    POSITIVE LOGITS
    ³³³
    1.01
    ³³³³³³³³³³³³³³³³
    0.94
    ³³³³
    0.88
    pmwiki
    0.85
    ³³³³³³³³
    0.82
    Beta
    0.82
    âĵĺ
    0.79
    Synopsis
    0.78
    Posted
    0.77
    Ingredients
    0.75
    Activations Density 1.667%

    No Known Activations