© 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 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. 6199
    Prev
    Next
    INDEX
    Explanations

    phrases related to news reporting, specifically mentioning locations and people

    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
    213
    +0.11
    0.5%
    365
    +0.11
    0.5%
    367
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    213
    +0.11
    0.03
    365
    +0.11
    0.03
    367
    +0.10
    0.03
    Negative Logits
    ãĥ¼ãĥĨãĤ£
    -0.81
    ãģį
    -0.73
    DragonMagazine
    -0.69
    å£
    -0.68
     Lovecraft
    -0.67
     sterling
    -0.67
    manship
    -0.66
    olicy
    -0.66
    lihood
    -0.65
    åij
    -0.65
    POSITIVE LOGITS
    manac
    1.30
    gebra
    1.21
    umni
    1.20
    ameda
    1.20
     Jazeera
    1.11
    gorithm
    1.10
    addin
    1.10
    uminium
    1.08
    oha
    1.06
    gorith
    1.06
    Activations Density 0.021%

    No Known Activations