© 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. 22167
    Prev
    Next
    INDEX
    Explanations

    academic degrees and educational qualifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    academic credentials and degrees

    oai_token-act-pair · gpt-4o-miniTriggered 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
    359
    +0.10
    0.5%
    302
    +0.10
    0.5%
    752
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    302
    +0.10
    0.03
    359
    +0.10
    0.03
    603
    +0.09
    0.03
    Negative Logits
     TMZ
    -0.80
     caravan
    -0.66
    ickr
    -0.65
     frontman
    -0.65
     bandwagon
    -0.65
     escal
    -0.64
     contag
    -0.64
     raids
    -0.64
     hypoc
    -0.64
     Clan
    -0.64
    POSITIVE LOGITS
    ploma
    1.38
     dissertation
    1.35
     Degree
    1.22
    ertation
    1.22
    inguished
    1.21
     diploma
    1.19
     doctoral
    1.17
     thesis
    1.15
     undergraduate
    1.10
    graduate
    1.10
    Activations Density 0.144%

    No Known Activations