© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 17219
    Prev
    Next
    INDEX
    Explanations

    phrases related to change or transition

    oai_token-act-pair · gpt-3.5-turbo

    repetitions of the conjunction "and."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Powered
    -0.74
    éŃĶ
    -0.70
    odcast
    -0.67
    Sup
    -0.63
     Doodle
    -0.63
    IOR
    -0.62
    Times
    -0.62
    è»
    -0.62
    Morning
    -0.62
    iors
    -0.59
    POSITIVE LOGITS
     therefore
    1.21
     consequently
    1.09
    rogen
    1.05
    rogens
    0.96
     hence
    0.96
     thus
    0.95
     then
    0.95
     possibly
    0.92
     secondly
    0.91
     thereby
    0.87
    Activations Density 0.802%

    No Known Activations