© 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. 1556
    Prev
    Next
    INDEX
    Explanations

    references to specific orders, rankings, or titles

    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
    137
    +0.15
    0.7%
    375
    +0.10
    0.5%
    425
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    137
    +0.15
    0.02
    425
    +0.10
    0.02
    97
    +0.09
    0.02
    Negative Logits
    gow
    -0.85
    iov
    -0.84
    iland
    -0.77
    zinski
    -0.70
    itions
    -0.69
    hunter
    -0.67
    nis
    -0.66
    SQL
    -0.65
    runtime
    -0.63
    igun
    -0.63
    POSITIVE LOGITS
     magnitude
    1.26
     precedence
    1.14
     priority
    0.82
     descent
    0.77
     elimination
    0.71
     nature
    0.71
     excellence
    0.71
    rontal
    0.70
     redundancy
    0.66
     succession
    0.66
    Activations Density 0.066%

    No Known Activations