© 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. 15961
    Prev
    Next
    INDEX
    Explanations

    phrases related to difficult or unfortunate circumstances

    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
    616
    +0.11
    0.5%
    408
    +0.11
    0.5%
    238
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    238
    +0.11
    0.03
    234
    +0.11
    0.03
    669
    +0.11
    0.02
    Negative Logits
    raq
    -0.80
    late
    -0.79
    çĶ°
    -0.77
    elled
    -0.74
    éĥ
    -0.72
    azel
    -0.71
    =-=-=-=-=-=-=-=-
    -0.68
    catentry
    -0.68
    soon
    -0.67
    ocations
    -0.67
    POSITIVE LOGITS
     interruption
    0.87
     ever
    0.82
     necessarily
    0.79
     mentioning
    0.79
     knowing
    0.78
     regard
    0.78
     touching
    0.77
     compromising
    0.76
     bothering
    0.75
    lessly
    0.75
    Activations Density 0.028%

    No Known Activations