© 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. 16815
    Prev
    Next
    INDEX
    Explanations

    legal and judicial terms related to criminal cases

    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
    586
    +0.12
    0.5%
    202
    +0.12
    0.5%
    719
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.07
    575
    +0.12
    0.06
    762
    +0.11
    0.06
    Negative Logits
    çĶ°
    -0.79
    iosyn
    -0.77
    soDeliveryDate
    -0.74
    Els
    -0.71
    erent
    -0.69
     unden
    -0.67
    Rap
    -0.67
    aeus
    -0.66
    hap
    -0.66
     theirs
    -0.65
    POSITIVE LOGITS
     syndrome
    1.23
     fallacy
    0.85
     ratio
    0.83
     Syndrome
    0.82
     Ratio
    0.77
     etc
    0.70
     ratios
    0.70
     implies
    0.67
     (
    0.67
     clauses
    0.66
    Activations Density 0.350%

    No Known Activations