© 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. 7689
    Prev
    Next
    INDEX
    Explanations

    phrases related to responsibility or critique towards particular individuals

    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
    737
    +0.11
    0.5%
    46
    +0.10
    0.5%
    371
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    737
    +0.11
    0.03
    542
    +0.10
    0.03
    741
    +0.10
    0.02
    Negative Logits
    urations
    -0.60
     Esp
    -0.60
    Bron
    -0.58
    orth
    -0.58
    ibaba
    -0.58
     Membership
    -0.57
     Shutterstock
    -0.57
     Trinity
    -0.56
    tesque
    -0.56
    Louis
    -0.55
    POSITIVE LOGITS
     responsible
    0.89
    liest
    0.84
     deciding
    0.82
     happiest
    0.79
     who
    0.73
    abet
    0.72
    responsible
    0.72
     reacting
    0.71
     initiating
    0.69
     risking
    0.69
    Activations Density 0.098%

    No Known Activations