© 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. 22532
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to the concept of origin

    oai_token-act-pair · gpt-4o-miniTriggered 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
    +0.14
    0.6%
    634
    +0.13
    0.6%
    541
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    634
    +0.14
    0.02
    541
    +0.13
    0.02
    668
    +0.12
    0.01
    Negative Logits
     helicop
    -0.70
    Agg
    -0.68
     Eye
    -0.62
     absentee
    -0.61
    ĵĺ
    -0.61
    owl
    -0.61
    PDATE
    -0.60
    eneg
    -0.60
     antioxid
    -0.58
     Patton
    -0.57
    POSITIVE LOGITS
    ator
    1.19
    ators
    1.18
    ates
    1.11
    uating
    0.91
    uation
    0.82
    itious
    0.80
    uated
    0.80
    acy
    0.79
    ating
    0.78
    ories
    0.77
    Activations Density 0.019%

    No Known Activations