© 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. 18496
    Prev
    Next
    INDEX
    Explanations

    locations and details related to physical places

    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
    +0.22
    1.1%
    430
    +0.16
    0.8%
    314
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.22
    0.09
    6
    +0.16
    0.08
    640
    +0.14
    0.08
    Negative Logits
     extinct
    -0.82
     viable
    -0.71
     sustain
    -0.69
     punish
    -0.69
     consistency
    -0.68
     diminish
    -0.68
     priorit
    -0.67
     reliant
    -0.67
     durability
    -0.67
     expend
    -0.67
    POSITIVE LOGITS
     Called
    0.90
     Afterwards
    0.89
     Apparently
    0.87
     Didn
    0.84
     Upon
    0.84
     Immediately
    0.79
    Interview
    0.78
     Meeting
    0.76
     Initially
    0.75
     Shortly
    0.74
    Activations Density 0.431%

    No Known Activations