© 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 24576
    9. 8-RES_FS24576-JB
    11. 18945
    Prev
    Next
    INDEX
    Explanations

    compounds or words related to acceleration or prioritization

    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
    450
    +0.13
    0.6%
    491
    +0.12
    0.5%
    575
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.13
    0.01
    491
    +0.12
    0.01
    251
    +0.10
    0.01
    Negative Logits
    enegger
    -0.73
     TVs
    -0.68
     Nicotine
    -0.67
    corn
    -0.67
     Dome
    -0.66
    Reloaded
    -0.66
    cham
    -0.65
    Interstitial
    -0.64
    Fram
    -0.62
    tery
    -0.59
    POSITIVE LOGITS
    itions
    1.40
    ited
    1.37
    iency
    1.35
    itious
    1.27
    iencies
    1.26
    iences
    1.13
    iting
    1.10
    ition
    1.10
    ience
    1.09
    itional
    1.05
    Activations Density 0.018%

    No Known Activations