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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 3072
    9. 8-RES_FS3072-JB
    11. 209
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns preceding proper nouns

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    362
    +0.15
    0.7%
    92
    +0.13
    0.6%
    15
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.15
    0.15
    15
    +0.13
    0.15
    553
    +0.12
    0.11
    Negative Logits
    ocument
    -0.84
    daq
    -0.84
    itect
    -0.82
     largeDownload
    -0.80
    otin
    -0.77
    eers
    -0.76
    inctions
    -0.74
    uve
    -0.71
    hari
    -0.70
    ibaba
    -0.70
    POSITIVE LOGITS
     own
    2.00
     wife
    1.30
     father
    1.23
     predecessor
    1.22
     superiors
    1.21
     daughter
    1.21
     favourite
    1.20
     namesake
    1.20
     newfound
    1.19
     predecessors
    1.19
    Activations Density 0.879%

    No Known Activations