© 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. 4246
    Prev
    Next
    INDEX
    Explanations

    items and objects that are not typically found in sewers, such as false teeth, a shower curtain, and a grenade

    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
    591
    +0.11
    0.5%
    283
    +0.10
    0.5%
    298
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    711
    +0.11
    0.04
    591
    +0.10
    0.05
    475
    +0.10
    0.06
    Negative Logits
    ":[{"
    -0.58
     guiName
    -0.53
    HOU
    -0.52
    odcast
    -0.47
    orah
    -0.47
    ERG
    -0.46
    hang
    -0.46
    verages
    -0.46
    resents
    -0.46
    Published
    -0.45
    POSITIVE LOGITS
    ?).
    2.09
    ').
    2.06
    !).
    2.06
    .).
    2.03
    )."
    2.00
    .)
    1.99
    ").
    1.96
     .)
    1.93
    .")
    1.92
    ]).
    1.91
    Activations Density 1.645%

    No Known Activations