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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 4078
    Prev
    Next
    INDEX
    Explanations

    phrases related to having a lot to do or discuss

    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
    696
    +0.11
    0.5%
    495
    +0.10
    0.5%
    397
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    495
    +0.11
    0.02
    591
    +0.10
    0.02
    650
    +0.10
    0.03
    Negative Logits
    pmwiki
    -0.84
    esson
    -0.74
    pecially
    -0.68
    mon
    -0.67
     coincides
    -0.64
    irection
    -0.60
    ounters
    -0.59
     Advertisement
    -0.59
     unsurprisingly
    -0.58
    lers
    -0.58
    POSITIVE LOGITS
    learn
    0.76
     learn
    0.73
     glean
    0.72
    izons
    0.69
     overcome
    0.67
     prove
    0.67
    KEN
    0.66
     unanswered
    0.65
    ðŁij
    0.65
     explore
    0.64
    Activations Density 0.079%

    No Known Activations