© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 45463
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "strip" or variations of it, suggesting a focus on topics related to stripping or the concept of being stripped down

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    +0.13
    0.6%
    315
    +0.12
    0.5%
    328
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    315
    +0.13
    0.01
    328
    +0.12
    0.01
    326
    +0.10
    0.01
    Negative Logits
    æŃ¦
    -0.80
    CLASSIFIED
    -0.80
    ORGE
    -0.73
    çĶ
    -0.69
     Gund
    -0.69
     Frey
    -0.69
    é¾
    -0.67
    ãĤ¢ãĥ«
    -0.67
    æĺ¯
    -0.67
    ãĥ¯ãĥ³
    -0.66
    POSITIVE LOGITS
    pling
    1.31
    ving
    1.24
    pper
    1.23
    ppers
    1.21
    pping
    1.20
    pped
    1.16
    ker
    1.12
    ppings
    1.11
    plings
    1.05
    ved
    1.02
    Activations Density 0.003%

    No Known Activations