© 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. 17025
    Prev
    Next
    INDEX
    Explanations

    terms related to the progress, implementation, structure, funding, situation, threat, pace of change, and tools in a technical context

    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
    447
    +0.20
    0.9%
    72
    +0.11
    0.5%
    288
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.20
    0.05
    225
    +0.11
    0.03
    575
    +0.09
    0.04
    Negative Logits
    DERR
    -0.71
    imb
    -0.64
    later
    -0.58
    sung
    -0.58
     later
    -0.56
    ammed
    -0.56
     Sat
    -0.56
    edi
    -0.56
    roit
    -0.55
     assassinated
    -0.55
    POSITIVE LOGITS
    ologies
    0.78
    bernatorial
    0.76
     paradigm
    0.74
     parad
    0.74
     incarnation
    0.73
     trajectory
    0.71
     iteration
    0.70
    Reviewer
    0.68
     trends
    0.66
    cture
    0.64
    Activations Density 0.217%

    No Known Activations