© 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 EvalsExportsAPI 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. 7544
    Prev
    Next
    INDEX
    Explanations

    information related to trends, statistics, and societal changes

    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
    763
    +0.12
    0.5%
    759
    +0.10
    0.5%
    691
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    763
    +0.12
    0.04
    691
    +0.10
    0.04
    324
    +0.10
    0.03
    Negative Logits
    apest
    -0.68
    ura
    -0.67
    scribe
    -0.66
    oss
    -0.63
    efer
    -0.60
    pei
    -0.59
    boro
    -0.59
     Talks
    -0.58
    mins
    -0.58
    erb
    -0.57
    POSITIVE LOGITS
     partly
    0.95
     factors
    0.93
     innate
    0.90
     exacerbated
    0.89
     attributable
    0.87
     stemming
    0.85
    Factor
    0.84
     inexper
    0.84
     generational
    0.83
     coincidence
    0.81
    Activations Density 0.201%

    No Known Activations