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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1398
    Prev
    Next
    INDEX
    Explanations

    mentions of people's names with the letter 'R' and 'M'

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    213
    +0.10
    0.5%
    496
    +0.10
    0.4%
    25
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.10
    0.05
    482
    +0.10
    0.05
    363
    +0.10
    0.05
    Negative Logits
    ñ
    -0.72
    20439
    -0.71
    ò
    -0.70
    ItemTracker
    -0.69
     pione
    -0.69
    DeliveryDate
    -0.66
     Parables
    -0.66
     eleph
    -0.65
     earthqu
    -0.63
    soDeliveryDate
    -0.62
    POSITIVE LOGITS
    .,
    0.91
    .?
    0.91
    .
    0.89
    .;
    0.85
    .-
    0.81
    ./
    0.77
    .:
    0.75
    .,"
    0.74
    ._
    0.73
    .–
    0.71
    Activations Density 0.768%

    No Known Activations