© 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 1536
    9. 8-RES_FS1536-JB
    11. 1076
    Prev
    Next
    INDEX
    Explanations

    language related to discussions, descriptions, and explanations of situations, decisions, and events

    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
    507
    +0.11
    0.5%
    320
    +0.11
    0.5%
    161
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    507
    +0.11
    0.10
    161
    +0.11
    0.06
    642
    +0.11
    0.06
    Negative Logits
     confir
    -0.70
    iem
    -0.66
    depth
    -0.64
     promoter
    -0.61
    sonian
    -0.61
     reached
    -0.58
     gathers
    -0.57
    ipeg
    -0.56
     complied
    -0.55
    iasm
    -0.54
    POSITIVE LOGITS
    phas
    1.05
    enance
    0.94
     themselves
    0.78
    igate
    0.76
    igated
    0.76
     ourselves
    0.75
     favorably
    0.72
    igating
    0.71
     blame
    0.67
     virtues
    0.66
    Activations Density 2.632%

    No Known Activations