© 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. 37408
    Prev
    Next
    INDEX
    Explanations

    elements associated with social media interactions and profiles

    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
    480
    +0.83
    7.2%
    266
    +0.19
    1.6%
    314
    +0.06
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.83
    -0.03
    6
    +0.19
    0.07
    480
    +0.06
    0.01
    Negative Logits
     losers
    -0.41
     secretaries
    -0.40
    wagen
    -0.39
     suites
    -0.39
     perfected
    -0.38
     paradise
    -0.36
     fundamentals
    -0.36
     successors
    -0.36
     plummeted
    -0.35
     adversaries
    -0.35
    POSITIVE LOGITS
    Aug
    0.44
    doi
    0.43
    19
    0.42
    UTH
    0.40
    59
    0.40
    1959
    0.40
    12
    0.39
     08
    0.39
    28
    0.39
    31
    0.39
    Activations Density 0.408%

    No Known Activations