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. 2460
    Prev
    Next
    INDEX
    Explanations

    references to Marvel comics and characters

    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
    701
    +0.10
    0.5%
    549
    +0.10
    0.5%
    14
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.09
    326
    +0.10
    0.07
    519
    +0.10
    0.06
    Negative Logits
    ntil
    -0.83
    ateur
    -0.77
    perate
    -0.72
    vant
    -0.72
    ickr
    -0.71
    mination
    -0.71
    uckland
    -0.71
    xual
    -0.71
    pex
    -0.70
    holm
    -0.69
    POSITIVE LOGITS
     Comics
    1.39
     Comic
    1.08
     Characters
    1.07
     Universe
    1.05
     villains
    1.05
     comics
    1.02
     Adventures
    1.02
     villain
    1.02
     superheroes
    1.00
     Animated
    0.93
    Activations Density 0.975%

    No Known Activations