© 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. 18539
    Prev
    Next
    INDEX
    Explanations

    references to the 9/11 attacks and related conspiracy theories

    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
    115
    +0.10
    0.5%
    127
    +0.10
    0.4%
    239
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    239
    +0.10
    0.04
    153
    +0.10
    0.04
    115
    +0.10
    0.04
    Negative Logits
    itably
    -0.75
    iating
    -0.71
    iated
    -0.71
    holders
    -0.70
    ivari
    -0.70
    iator
    -0.67
    itable
    -0.67
    enture
    -0.66
    pmwiki
    -0.66
    iations
    -0.65
    POSITIVE LOGITS
    9999
    1.36
    999
    1.27
    06
    1.22
    090
    1.20
    07
    1.14
    08
    1.12
    03
    1.10
    04
    1.08
    02
    1.04
    09
    1.02
    Activations Density 0.065%

    No Known Activations