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. 598
    Prev
    Next
    INDEX
    Explanations

    comparisons and analogies using the word "like" followed by a comparison to physical or abstract concepts

    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
    75
    +0.12
    0.5%
    90
    +0.09
    0.4%
    105
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    75
    +0.12
    0.03
    692
    +0.09
    0.03
    105
    +0.09
    0.03
    Negative Logits
    erity
    -0.79
    iencies
    -0.78
    ixel
    -0.76
    folios
    -0.72
    rity
    -0.71
    Published
    -0.71
    iyah
    -0.70
    papers
    -0.70
    idates
    -0.70
    otte
    -0.69
    POSITIVE LOGITS
    lier
    1.10
    liest
    1.06
    lihood
    0.89
     waking
    0.80
     heaven
    0.78
     spitting
    0.77
     comparing
    0.74
     wildfire
    0.73
     slap
    0.71
     inverse
    0.70
    Activations Density 0.076%

    No Known Activations