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. 2120
    Prev
    Next
    INDEX
    Explanations

    references to news articles and statements about a person's death

    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
    13
    +0.10
    0.4%
    263
    +0.10
    0.4%
    105
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    105
    +0.10
    0.08
    13
    +0.10
    0.09
    624
    +0.09
    0.05
    Negative Logits
     stuff
    -0.78
     weed
    -0.73
     cram
    -0.71
     lurking
    -0.70
     looms
    -0.68
    stuff
    -0.67
     tended
    -0.67
     stash
    -0.67
     fert
    -0.67
     filler
    -0.67
    POSITIVE LOGITS
     saddened
    1.25
     THANK
    1.16
     condolences
    1.14
     hereby
    1.14
     unres
    1.13
    Thank
    1.12
     congratulate
    1.08
     thank
    1.07
     congratulations
    1.06
     Thank
    1.06
    Activations Density 2.639%

    No Known Activations