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. 2360
    Prev
    Next
    INDEX
    Explanations

    mentions of sporting events, particularly hockey and basketball

    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
    281
    +0.15
    0.7%
    430
    +0.10
    0.5%
    603
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    281
    +0.15
    0.10
    238
    +0.10
    0.08
    603
    +0.10
    0.07
    Negative Logits
    ickr
    -0.94
     lawy
    -0.88
     informants
    -0.83
     memos
    -0.79
    ework
    -0.78
     writings
    -0.77
     manuals
    -0.77
    umblr
    -0.76
     handwriting
    -0.74
     Pastebin
    -0.74
    POSITIVE LOGITS
     undefeated
    1.28
     Finals
    1.26
     rematch
    1.26
     championship
    1.25
     Championship
    1.23
     Stadium
    1.20
     semifinals
    1.16
     winner
    1.16
     finals
    1.16
     Coliseum
    1.16
    Activations Density 3.133%

    No Known Activations