© 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. 21097
    Prev
    Next
    INDEX
    Explanations

    phrases related to effective communication and collaboration in a work setting

    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
    +0.20
    0.9%
    447
    +0.15
    0.7%
    170
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.20
    0.05
    345
    +0.15
    0.03
    669
    +0.11
    0.03
    Negative Logits
    ©¶æ
    -0.79
    agues
    -0.73
    wic
    -0.71
    ãĤ§
    -0.71
    cler
    -0.70
    cffffcc
    -0.70
    Semitic
    -0.70
    Sem
    -0.70
    ä
    -0.70
    VI
    -0.70
    POSITIVE LOGITS
     patience
    1.06
     ensuring
    1.06
     maintaining
    1.01
     educating
    1.01
     overcoming
    0.96
     knowing
    0.96
     remembering
    0.94
     preserving
    0.93
     protecting
    0.90
     mastering
    0.90
    Activations Density 0.187%

    No Known Activations