© 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. 10639
    Prev
    Next
    INDEX
    Explanations

    question marks at the end of sentences

    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
    480
    +0.18
    0.8%
    447
    +0.10
    0.5%
    166
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    285
    +0.18
    0.03
    6
    +0.10
    0.04
    166
    +0.08
    0.03
    Negative Logits
    ulla
    -0.65
     shortest
    -0.64
    outine
    -0.64
    ened
    -0.63
     Nadu
    -0.62
    aper
    -0.61
     Marshal
    -0.60
    orough
    -0.59
    Hour
    -0.59
     antip
    -0.58
    POSITIVE LOGITS
     Nope
    1.09
    ����
    0.98
     Seems
    0.96
     Wouldn
    0.96
    ¶
    0.95
     Surely
    0.93
     Yeah
    0.92
     Why
    0.89
     Certainly
    0.89
     Would
    0.86
    Activations Density 0.059%

    No Known Activations