© 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. 1193
    Prev
    Next
    INDEX
    Explanations

    news articles related to tragic events, accidents, and incidents

    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
    546
    +0.12
    0.5%
    689
    +0.10
    0.4%
    494
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    546
    +0.12
    0.10
    43
    +0.10
    0.07
    765
    +0.09
    0.07
    Negative Logits
    ²¾
    -0.75
    ocl
    -0.73
    enth
    -0.72
    sburg
    -0.71
    uten
    -0.68
    estones
    -0.67
    oph
    -0.64
    cloth
    -0.63
    geist
    -0.63
    lihood
    -0.62
    POSITIVE LOGITS
     according
    1.01
     prompting
    0.98
     sparking
    0.94
     including
    0.83
     CNBC
    0.82
     POLITICO
    0.81
    according
    0.78
     marking
    0.77
    AFP
    0.77
     deepening
    0.75
    Activations Density 0.436%

    No Known Activations