© 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 3072
    9. 8-RES_FS3072-JB
    11. 151
    Prev
    Next
    INDEX
    Explanations

    dates mentioned in news articles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    509
    +0.13
    0.6%
    50
    +0.10
    0.5%
    26
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.13
    0.10
    62
    +0.10
    0.08
    26
    +0.10
    0.08
    Negative Logits
    itiz
    -0.64
    rouse
    -0.59
    :/
    -0.59
     tresp
    -0.59
     choke
    -0.56
     damned
    -0.55
     spoil
    -0.55
     offensive
    -0.54
     whore
    -0.53
     professionally
    -0.53
    POSITIVE LOGITS
    ward
    1.06
    neath
    0.96
     contrast
    0.91
    itialized
    0.87
    cest
    0.86
    ventory
    0.85
    coming
    0.85
    wards
    0.84
    juries
    0.81
    look
    0.81
    Activations Density 0.669%

    No Known Activations