© 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. 7416
    Prev
    Next
    INDEX
    Explanations

    information and discussions about accuracy, specifically related to data and estimates

    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
    7
    +0.12
    0.5%
    140
    +0.10
    0.5%
    448
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    455
    +0.12
    0.02
    162
    +0.10
    0.01
    140
    +0.10
    0.02
    Negative Logits
    hedon
    -0.73
    hov
    -0.70
    horn
    -0.70
    doms
    -0.69
    AIN
    -0.69
    wart
    -0.68
    ATA
    -0.68
     Pharma
    -0.67
    ovan
    -0.66
    inki
    -0.66
    POSITIVE LOGITS
     portrayal
    1.12
     depiction
    1.02
    uracy
    0.96
     reproduction
    0.93
     representation
    0.91
     reprodu
    0.89
     depictions
    0.89
     appraisal
    0.87
     description
    0.86
     portray
    0.85
    Activations Density 0.029%

    No Known Activations