© 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 12288
    9. 8-RES_FS12288-JB
    11. 790
    Prev
    Next
    INDEX
    Explanations

    websites and sources

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    175
    +0.14
    0.7%
    545
    +0.11
    0.5%
    679
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    175
    +0.14
    0.05
    679
    +0.11
    0.07
    345
    +0.11
    0.06
    Negative Logits
    ties
    -0.69
    plane
    -0.69
    lier
    -0.68
    ALLY
    -0.66
    alysis
    -0.64
    ously
    -0.64
    oÄŁ
    -0.64
    aways
    -0.63
    eur
    -0.60
    planes
    -0.60
    POSITIVE LOGITS
    ource
    1.40
    ystem
    1.29
    peed
    1.25
    heet
    1.23
    mith
    1.23
    cript
    1.22
    pace
    1.18
    hip
    1.18
    hare
    1.17
    ilver
    1.14
    Activations Density 0.453%

    No Known Activations