© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 41786
    Prev
    Next
    INDEX
    Explanations

    references to cloud storage services, particularly Dropbox and iCloud

    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
    4
    +0.12
    0.5%
    346
    +0.12
    0.5%
    683
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    343
    +0.12
    0.01
    4
    +0.12
    0.01
    294
    +0.12
    0.01
    Negative Logits
    stad
    -0.80
    sein
    -0.76
    66666666
    -0.73
    ammad
    -0.73
    colm
    -0.72
    abad
    -0.71
    eering
    -0.70
    <<
    -0.68
    omaly
    -0.67
    atown
    -0.67
    POSITIVE LOGITS
     Dropbox
    1.00
     Sync
    0.98
    Drop
    0.89
     folders
    0.80
     storage
    0.78
    pedia
    0.76
    files
    0.76
     upload
    0.75
     servers
    0.74
     synchronization
    0.73
    Activations Density 0.010%

    No Known Activations