© 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 1536
    9. 8-RES_FS1536-JB
    11. 99
    Prev
    Next
    INDEX
    Explanations

    text related to computer programming languages, specifically focusing on the syntax and components of the language

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    192
    +0.10
    0.5%
    647
    +0.10
    0.5%
    393
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    647
    +0.10
    0.12
    172
    +0.10
    0.12
    441
    +0.10
    0.10
    Negative Logits
    oller
    -0.69
    ople
    -0.67
    eor
    -0.61
    itiveness
    -0.60
    audi
    -0.58
    osc
    -0.58
    Rub
    -0.58
    romy
    -0.57
    rolet
    -0.57
    quist
    -0.56
    POSITIVE LOGITS
     opposed
    1.25
    pired
    1.24
    pires
    1.18
     well
    1.17
     evidenced
    1.14
    ynchron
    1.10
    phy
    1.10
     follows
    1.09
    piring
    1.03
    pects
    1.01
    Activations Density 0.695%

    No Known Activations