© 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. 1026
    Prev
    Next
    INDEX
    Explanations

    expressions of approval or praise, particularly with the phrase "not only."

    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
    373
    +0.14
    0.7%
    458
    +0.11
    0.5%
    717
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    458
    +0.14
    0.06
    345
    +0.11
    0.05
    184
    +0.10
    0.03
    Negative Logits
    »Ĵ
    -0.89
    nect
    -0.68
     nonetheless
    -0.65
    etc
    -0.64
    ³
    -0.61
    ashington
    -0.60
    Ŀ
    -0.60
    EStream
    -0.59
     Frenzy
    -0.59
    ues
    -0.59
    POSITIVE LOGITS
    ifiable
    0.77
    oped
    0.68
    tem
    0.67
     physically
    0.66
    ional
    0.60
     obe
    0.60
    onia
    0.58
    lihood
    0.58
     combatants
    0.57
    onne
    0.56
    Activations Density 1.201%

    No Known Activations