© 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 12288
    9. 8-RES_FS12288-JB
    11. 1706
    Prev
    Next
    INDEX
    Explanations

    references to assistance or aid

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    679
    +0.17
    0.8%
    253
    +0.16
    0.7%
    458
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    253
    +0.17
    0.03
    288
    +0.16
    0.03
    379
    +0.09
    0.02
    Negative Logits
    place
    -0.76
    aval
    -0.69
    nen
    -0.66
    nces
    -0.65
    iple
    -0.64
    brace
    -0.64
    tail
    -0.63
    ndra
    -0.61
    agall
    -0.60
    pool
    -0.60
    POSITIVE LOGITS
     Desk
    0.86
     aid
    0.81
     assisting
    0.75
    assisted
    0.71
    itech
    0.70
     assistance
    0.70
    sie
    0.69
    des
    0.68
    abet
    0.67
    ilitation
    0.66
    Activations Density 0.071%

    No Known Activations