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 49152
    5. 8-RES_FS49152-JB
    6. 40102
    Prev
    Next
    INDEX
    Explanations

    calls for assistance or support, particularly in the context of ongoing issues or actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    534
    +0.11
    0.5%
    740
    +0.09
    0.4%
    449
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.11
    0.03
    181
    +0.09
    0.03
    211
    +0.09
    0.01
    Negative Logits
     monop
    -0.80
    ModLoader
    -0.75
     mathemat
    -0.74
    20439
    -0.73
    è¦ļéĨĴ
    -0.72
    ãģ®å®
    -0.69
    åĮ
    -0.68
    åŃ
    -0.68
    mini
    -0.68
    oother
    -0.68
    POSITIVE LOGITS
    isons
    0.86
    gers
    0.79
    esters
    0.78
    berman
    0.75
    és
    0.75
    pires
    0.74
    ester
    0.72
    pired
    0.72
    velt
    0.72
     us
    0.71
    Activations Density 0.085%

    No Known Activations