© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 768
    9. 8-RES_FS768-JB
    11. 715
    Prev
    Next
    INDEX
    Explanations

    mentions of car accidents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    89
    +0.11
    0.5%
    652
    +0.10
    0.5%
    593
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.18
    537
    +0.10
    0.13
    652
    +0.10
    0.12
    Negative Logits
     Seym
    -0.75
    arial
    -0.72
     Flavoring
    -0.71
    ordial
    -0.69
    vironment
    -0.68
    ertain
    -0.67
     Lum
    -0.67
    xus
    -0.66
    aspberry
    -0.66
    achu
    -0.66
    POSITIVE LOGITS
     motorcycles
    1.21
     cars
    1.19
    bike
    1.17
     racing
    1.17
     driver
    1.14
     drivers
    1.13
     parked
    1.07
    wagon
    1.06
     racer
    1.05
    cars
    1.05
    Activations Density 4.365%

    No Known Activations