© 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 24576
    9. 8-RES_FS24576-JB
    11. 2249
    Prev
    Next
    INDEX
    Explanations

    phrases related to crime investigation and detection

    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
    314
    +0.17
    0.8%
    651
    +0.14
    0.6%
    311
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.17
    0.04
    288
    +0.14
    0.04
    314
    +0.11
    0.00
    Negative Logits
     deaf
    -0.69
    FORE
    -0.68
     fres
    -0.62
    agine
    -0.60
     nerv
    -0.60
     awake
    -0.60
    zzle
    -0.59
    butt
    -0.59
     assurance
    -0.59
     ashore
    -0.58
    POSITIVE LOGITS
    ions
    1.84
    ive
    1.63
    ives
    1.62
    ivity
    1.57
    ively
    1.51
    ors
    1.51
    ing
    1.38
    ivist
    1.36
    orate
    1.36
    s
    1.35
    Activations Density 0.108%

    No Known Activations