© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 4579
    Prev
    Next
    INDEX
    Explanations

    technical descriptions or reporting of accidents/incidents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.51
    2.7%
    266
    +0.11
    0.6%
    614
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    482
    +0.51
    0.04
    181
    +0.11
    0.04
    153
    +0.09
    0.04
    Negative Logits
     Walters
    -0.57
     Hodg
    -0.57
     Guatem
    -0.53
    Haw
    -0.53
     Hob
    -0.51
     Garner
    -0.49
    Dur
    -0.49
    DK
    -0.49
    Thompson
    -0.49
     Din
    -0.49
    POSITIVE LOGITS
    actionDate
    0.69
    utterstock
    0.65
     Flavoring
    0.61
     largeDownload
    0.59
    OTOS
    0.58
    Reviewer
    0.57
    ollah
    0.55
    icago
    0.55
     nig
    0.55
    ushima
    0.54
    Activations Density 0.320%

    No Known Activations