© 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. 9114
    Prev
    Next
    INDEX
    Explanations

    names of movies, episode titles, and specific details related to film and television

    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
    481
    +0.16
    0.7%
    447
    +0.12
    0.5%
    762
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.16
    0.07
    291
    +0.12
    0.06
    247
    +0.11
    0.05
    Negative Logits
    edu
    -0.62
    uld
    -0.60
    userc
    -0.59
    oren
    -0.57
     archives
    -0.57
    ceivable
    -0.57
     resolutions
    -0.57
    rences
    -0.56
     Dreams
    -0.56
     histories
    -0.56
    POSITIVE LOGITS
     extraord
    1.08
    agonist
    0.79
     overseeing
    0.76
     guarding
    0.76
     archetype
    0.76
    izer
    0.75
     alongside
    0.75
     unto
    0.74
    iment
    0.74
     handler
    0.73
    Activations Density 0.308%

    No Known Activations