© 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. 20255
    Prev
    Next
    INDEX
    Explanations

    phrases expressing affirmative agreement

    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
    255
    +0.12
    0.6%
    572
    +0.11
    0.5%
    89
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.12
    0.02
    119
    +0.11
    0.01
    273
    +0.10
    0.02
    Negative Logits
    tein
    -0.87
    RAW
    -0.75
    rament
    -0.73
    bable
    -0.69
    portion
    -0.67
    achable
    -0.67
    actionDate
    -0.67
    joined
    -0.64
    IENT
    -0.63
    minist
    -0.63
    POSITIVE LOGITS
    terday
    1.28
    yeah
    1.16
     Yeah
    1.15
     yeah
    1.10
    hhhh
    1.06
    hhh
    0.99
     kidding
    0.96
    hh
    0.93
    Yeah
    0.86
     thats
    0.82
    Activations Density 0.019%

    No Known Activations