© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 22095
    Prev
    Next
    INDEX
    Explanations

    forms of the word "succeed" or variations indicating success

    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
    631
    +0.12
    0.6%
    193
    +0.10
    0.4%
    745
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    631
    +0.12
    0.02
    641
    +0.10
    0.02
    12
    +0.10
    0.01
    Negative Logits
     Ferdinand
    -0.65
     Weir
    -0.61
     union
    -0.59
     sinks
    -0.59
     Orders
    -0.59
    wright
    -0.58
     Sinclair
    -0.57
     shorts
    -0.56
    ãĤº
    -0.56
     Noir
    -0.56
    POSITIVE LOGITS
    zanne
    1.28
    ction
    1.19
    nder
    1.19
    fficiency
    1.19
    ccess
    1.18
    cess
    1.10
    cc
    1.06
    prisingly
    1.05
    perm
    1.01
    ctions
    1.00
    Activations Density 0.015%

    No Known Activations