© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 45654
    Prev
    Next
    INDEX
    Explanations

    subordinate clauses beginning with "that"

    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
    507
    +0.13
    0.6%
    447
    +0.11
    0.5%
    393
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    601
    +0.13
    0.03
    203
    +0.11
    0.03
    228
    +0.10
    0.03
    Negative Logits
    backer
    -0.95
    Detailed
    -0.75
    Guard
    -0.72
    stal
    -0.71
    anie
    -0.68
    horn
    -0.68
    cellaneous
    -0.67
    rawdownloadcloneembedreportprint
    -0.66
    Details
    -0.66
    details
    -0.66
    POSITIVE LOGITS
     someday
    0.89
    thood
    0.86
     homosexuality
    0.84
     humans
    0.77
     rationality
    0.76
     immutable
    0.75
     mankind
    0.74
     there
    0.74
     sexuality
    0.73
     morality
    0.72
    Activations Density 0.078%

    No Known Activations