© 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. 21144
    Prev
    Next
    INDEX
    Explanations

    phrases related to commentary on current events or political situations

    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
    756
    +0.13
    0.6%
    720
    +0.11
    0.5%
    71
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.13
    -0.02
    703
    +0.11
    0.03
    156
    +0.10
    0.03
    Negative Logits
    xit
    -0.74
    hens
    -0.65
     IPM
    -0.64
     Variant
    -0.63
    illas
    -0.62
     Because
    -0.61
     Join
    -0.61
    Newsletter
    -0.60
    ohm
    -0.59
    fixes
    -0.59
    POSITIVE LOGITS
     already
    0.94
     rarely
    0.85
     notoriously
    0.80
     technically
    0.73
     normally
    0.73
     only
    0.73
     previous
    0.72
     often
    0.72
     practically
    0.72
     frequently
    0.72
    Activations Density 0.662%

    No Known Activations