© 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. 16005
    Prev
    Next
    INDEX
    Explanations

    conjunctions followed by contrasting information or ideas

    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
    393
    +0.12
    0.6%
    8
    +0.12
    0.6%
    203
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.12
    0.04
    8
    +0.12
    0.04
    598
    +0.11
    0.03
    Negative Logits
    asks
    -0.71
    ct
    -0.69
    meta
    -0.67
    wake
    -0.64
    coat
    -0.63
    ILLE
    -0.63
    nce
    -0.60
    illary
    -0.59
    endar
    -0.58
    orc
    -0.57
    POSITIVE LOGITS
     rather
    1.34
    rather
    1.10
     merely
    1.08
     simply
    0.97
     Rather
    0.92
     instead
    0.89
     suffice
    0.87
     nevertheless
    0.84
     nonetheless
    0.75
    Rather
    0.75
    Activations Density 0.054%

    No Known Activations