© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1093
    Prev
    Next
    INDEX
    Explanations

    book titles and movie names

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    232
    +0.14
    0.6%
    6
    +0.10
    0.5%
    476
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.14
    0.17
    145
    +0.10
    0.10
    476
    +0.10
    0.07
    Negative Logits
     customary
    -0.81
     parity
    -0.79
     referen
    -0.78
     theoretically
    -0.78
     deleg
    -0.76
     tert
    -0.76
    umerous
    -0.75
     terminology
    -0.75
     podium
    -0.74
     preventive
    -0.74
    POSITIVE LOGITS
     Dreams
    1.27
     Lies
    1.23
     Ruin
    1.21
     Lives
    1.21
     Bastard
    1.21
     Skies
    1.20
     Secrets
    1.17
     Forever
    1.16
     Madness
    1.15
     Angels
    1.15
    Activations Density 2.650%

    No Known Activations