Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18973
    Prev
    Next
    INDEX
    Explanations

    verbs related to requesting assistance or summoning, typically in a dramatic or urgent context

    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
    286
    +0.13
    0.6%
    755
    +0.13
    0.6%
    343
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.01
    286
    +0.13
    0.01
    343
    +0.12
    0.01
    Negative Logits
    hirt
    -0.90
    cas
    -0.76
    imb
    -0.72
    nut
    -0.70
    ijk
    -0.70
    inion
    -0.68
    cover
    -0.67
    pract
    -0.66
    sonian
    -0.66
    acca
    -0.66
    POSITIVE LOGITS
    eers
    0.91
     summon
    0.87
     summoning
    0.86
     demons
    0.83
    eer
    0.79
     Summon
    0.78
     summoned
    0.77
     minions
    0.77
    aurus
    0.77
     Cth
    0.76
    Activations Density 0.040%

    No Known Activations