© 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. 16272
    Prev
    Next
    INDEX
    Explanations

    references to physical actions or characteristics of objects, such as groping, grazed, and cheeses

    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
    526
    +0.14
    0.6%
    218
    +0.12
    0.5%
    380
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    588
    +0.14
    0.03
    326
    +0.12
    0.03
    526
    +0.10
    0.01
    Negative Logits
     LCS
    -0.70
     motto
    -0.68
     Bravo
    -0.66
    WAYS
    -0.66
     tyrann
    -0.65
    STAR
    -0.64
     pride
    -0.61
     Doctrine
    -0.61
     Yankee
    -0.61
     shift
    -0.60
    POSITIVE LOGITS
    es
    1.51
    ements
    1.37
    ement
    1.29
    ecake
    1.24
    esy
    1.13
    ername
    1.13
    er
    1.12
    ers
    1.12
    orial
    1.09
    ations
    1.06
    Activations Density 0.086%

    No Known Activations