© 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. 45619
    Prev
    Next
    INDEX
    Explanations

    phrases indicating missed opportunities or events

    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
    680
    +0.12
    0.5%
    172
    +0.10
    0.4%
    136
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    504
    +0.12
    0.01
    680
    +0.10
    0.01
    136
    +0.09
    0.01
    Negative Logits
    ysis
    -0.74
    vec
    -0.72
    seek
    -0.72
    rim
    -0.72
    onse
    -0.69
    venge
    -0.68
    dh
    -0.66
    rouse
    -0.65
    levels
    -0.65
    reth
    -0.64
    POSITIVE LOGITS
    pelled
    1.02
     something
    0.75
    uled
    0.73
    poke
    0.73
    pell
    0.73
     LAST
    0.71
    pelling
    0.71
     noticing
    0.70
     Notice
    0.65
     spotting
    0.64
    Activations Density 0.012%

    No Known Activations