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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 26667
    Prev
    Next
    INDEX
    Explanations

    various forms of communication about potential developments or rumors

    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
    447
    +0.19
    0.8%
    395
    +0.13
    0.6%
    411
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    395
    +0.19
    0.03
    411
    +0.13
    0.03
    195
    +0.10
    0.02
    Negative Logits
    Together
    -0.74
    oho
    -0.70
    alone
    -0.66
     Alone
    -0.66
    Experience
    -0.62
    Located
    -0.61
     Delicious
    -0.60
    alos
    -0.59
    oku
    -0.57
    nea
    -0.57
    POSITIVE LOGITS
     rumours
    0.82
     speculate
    0.80
     lately
    0.79
     suggesting
    0.78
     rumors
    0.77
     circulating
    0.77
     regarding
    0.77
     alleging
    0.75
     amongst
    0.72
     concerning
    0.70
    Activations Density 0.125%

    No Known Activations