© 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. 45483
    Prev
    Next
    INDEX
    Explanations

    references to specific time periods 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
    457
    +0.11
    0.5%
    129
    +0.10
    0.5%
    419
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    679
    +0.11
    0.05
    375
    +0.10
    0.02
    707
    +0.10
    0.03
    Negative Logits
     unspecified
    -0.72
    rogen
    -0.67
    Fax
    -0.67
    ertodd
    -0.64
     Frie
    -0.63
     yuan
    -0.61
    ager
    -0.61
     knob
    -0.60
    aith
    -0.58
     listener
    -0.58
    POSITIVE LOGITS
    where
    0.84
     whereby
    0.83
     where
    0.79
     wherein
    0.77
    ilogy
    0.67
    Reviewer
    0.67
    uggest
    0.66
     famously
    0.63
     tragedies
    0.62
    ¥ŀ
    0.61
    Activations Density 0.319%

    No Known Activations