© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6588
    Prev
    Next
    INDEX
    Explanations

    specific mentions of accounting or taking into account various factors and features

    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
    618
    +0.12
    0.6%
    71
    +0.11
    0.5%
    101
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    691
    +0.12
    0.02
    157
    +0.11
    0.02
    641
    +0.11
    0.02
    Negative Logits
    sis
    -0.84
    sil
    -0.79
    quer
    -0.76
    nect
    -0.75
    soever
    -0.74
    cit
    -0.73
    Constructed
    -0.72
    azes
    -0.69
    Torrent
    -0.67
    hero
    -0.64
    POSITIVE LOGITS
     factors
    1.08
     considerations
    1.03
    enance
    0.92
     probabilities
    0.88
     mitigating
    0.87
     how
    0.77
     biases
    0.75
     budgetary
    0.74
    ancy
    0.74
     factor
    0.73
    Activations Density 0.043%

    No Known Activations