© 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 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. 10991
    Prev
    Next
    INDEX
    Explanations

    information related to financial markets and legal issues

    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
    459
    +0.16
    0.8%
    756
    +0.14
    0.7%
    87
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.16
    -0.03
    224
    +0.14
    0.06
    410
    +0.12
    0.04
    Negative Logits
    à¨
    -0.63
     Whilst
    -0.62
    cult
    -0.61
    colour
    -0.61
     tremend
    -0.59
    âĺħ
    -0.59
    Written
    -0.58
     fucking
    -0.58
    Firstly
    -0.57
     civilisation
    -0.57
    POSITIVE LOGITS
     meanwhile
    1.00
     afterward
    0.91
     however
    0.89
     later
    0.87
     declined
    0.78
     also
    0.77
     reportedly
    0.77
     unsuccessfully
    0.76
     additionally
    0.75
     spokeswoman
    0.75
    Activations Density 1.295%

    No Known Activations