© 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. 17213
    Prev
    Next
    INDEX
    Explanations

    phrases related to decision-making and planning

    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
    167
    +0.15
    0.7%
    183
    +0.15
    0.7%
    26
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    183
    +0.15
    0.04
    26
    +0.15
    0.03
    276
    +0.10
    0.03
    Negative Logits
    nown
    -0.79
    ican
    -0.72
     bom
    -0.69
    onte
    -0.66
    Unix
    -0.64
    keep
    -0.63
    terday
    -0.60
    unker
    -0.60
    pine
    -0.58
    etheless
    -0.58
    POSITIVE LOGITS
     parallels
    0.97
    cards
    0.91
    strings
    0.89
    card
    0.87
     attention
    0.85
     inspiration
    0.84
     conclusions
    0.82
     drawn
    0.82
     Cosponsors
    0.80
     nearer
    0.80
    Activations Density 0.032%

    No Known Activations