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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1136
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to people or places

    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
    326
    +0.13
    0.6%
    453
    +0.12
    0.6%
    269
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.12
    288
    +0.12
    0.13
    464
    +0.11
    0.08
    Negative Logits
    phis
    -0.64
    orea
    -0.64
     Corpus
    -0.63
    terson
    -0.62
    asia
    -0.61
    alach
    -0.61
    SUP
    -0.61
    anooga
    -0.61
    mercial
    -0.61
     Speedway
    -0.60
    POSITIVE LOGITS
    senal
    1.31
    cliffe
    1.14
    utherford
    0.84
    MpServer
    0.83
    dies
    0.81
    ledge
    0.80
    ussia
    0.78
    aldo
    0.77
    abbit
    0.76
    Beer
    0.75
    Activations Density 1.547%

    No Known Activations