© 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. 16548
    Prev
    Next
    INDEX
    Explanations

    phrases related to travel and tourism, particularly focusing on community projects and cultural experiences

    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
    479
    +0.15
    0.7%
    391
    +0.13
    0.6%
    696
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.15
    0.04
    461
    +0.13
    0.04
    391
    +0.10
    0.03
    Negative Logits
    iris
    -0.89
    nos
    -0.89
    akia
    -0.87
    wal
    -0.85
    respond
    -0.84
    theless
    -0.84
    mos
    -0.84
    eyed
    -0.83
    bat
    -0.82
    headed
    -0.81
    POSITIVE LOGITS
     Collection
    1.07
     Community
    1.06
     Badge
    1.04
     Challenge
    1.04
     Contest
    1.01
     Festival
    1.00
     Quarterly
    1.00
     Program
    1.00
     Presents
    0.99
     Assist
    0.99
    Activations Density 0.227%

    No Known Activations