© 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 3072
    9. 8-RES_FS3072-JB
    11. 3052
    Prev
    Next
    INDEX
    Explanations

    phrases related to interaction and mutual experiences between individuals

    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
    528
    +0.10
    0.4%
    502
    +0.10
    0.4%
    506
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    528
    +0.10
    0.07
    153
    +0.10
    0.06
    304
    +0.09
    0.06
    Negative Logits
    duc
    -0.72
    phies
    -0.69
    arte
    -0.67
    pletion
    -0.66
    achers
    -0.65
    arah
    -0.65
    elight
    -0.64
    iries
    -0.63
    ervation
    -0.63
    kie
    -0.63
    POSITIVE LOGITS
     mutually
    1.37
     insepar
    1.14
     mutual
    1.03
     together
    1.02
     jointly
    1.00
     exchanging
    0.94
    together
    0.92
     agree
    0.90
     Together
    0.89
     respectively
    0.89
    Activations Density 1.684%

    No Known Activations