© 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. 6782
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to societal/political issues and perspectives

    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
    566
    +0.16
    0.7%
    663
    +0.13
    0.6%
    95
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    95
    +0.16
    0.04
    566
    +0.13
    0.04
    661
    +0.13
    0.04
    Negative Logits
     Xuan
    -0.64
     Trap
    -0.64
     Kag
    -0.63
     Gutenberg
    -0.62
     Hels
    -0.59
     HT
    -0.58
     Od
    -0.58
     Gad
    -0.58
     Hebdo
    -0.58
     Fortune
    -0.58
    POSITIVE LOGITS
     advise
    1.16
     recommend
    1.12
     prefer
    1.12
     sugg
    1.08
     gladly
    1.06
     suggest
    1.02
     characterize
    1.01
     appreciate
    1.00
     encourage
    0.99
     expect
    0.98
    Activations Density 0.065%

    No Known Activations