© 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. 18776
    Prev
    Next
    INDEX
    Explanations

    contractions of the verb 'are'

    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
    608
    +0.14
    0.6%
    519
    +0.13
    0.6%
    102
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    519
    +0.14
    0.06
    566
    +0.13
    0.06
    55
    +0.12
    0.06
    Negative Logits
    ESE
    -0.77
     Reduce
    -0.72
    DS
    -0.67
     mater
    -0.67
    TAIN
    -0.67
    andise
    -0.65
     membr
    -0.64
    ren
    -0.60
    iates
    -0.60
    iox
    -0.59
    POSITIVE LOGITS
     gonna
    1.50
     gotta
    1.10
     going
    1.06
     hoping
    1.04
     supposed
    1.00
     afraid
    0.97
     glad
    0.96
     guessing
    0.96
     not
    0.94
     sorry
    0.93
    Activations Density 0.064%

    No Known Activations