© 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. 2751
    Prev
    Next
    INDEX
    Explanations

    terms related to neurological and psychiatric research

    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
    640
    +0.26
    1.2%
    423
    +0.11
    0.5%
    615
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.26
    0.09
    160
    +0.11
    0.06
    64
    +0.11
    -0.02
    Negative Logits
     Unicorn
    -0.76
     Greg
    -0.71
     lett
    -0.70
     Wiz
    -0.70
     renting
    -0.70
     everywhere
    -0.70
     freel
    -0.70
     outgoing
    -0.68
     Healer
    -0.67
     flo
    -0.67
    POSITIVE LOGITS
    Introduction
    1.41
    Acknowled
    1.36
    Discussion
    1.36
    RESULTS
    1.34
    Abstract
    1.33
    METHOD
    1.31
    Methods
    1.26
    Figure
    1.23
    FIG
    1.21
    TABLE
    1.20
    Activations Density 0.204%

    No Known Activations