© 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. 8-RES-JB
    11. 12845
    Prev
    Next
    INDEX
    Explanations

    terms related to actions to be taken or decisions to be made

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a "must-have" quality or necessity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

    Negative Logits
     Arrow
    -0.72
     Rig
    -0.69
     Horn
    -0.66
     Vessel
    -0.65
     dart
    -0.64
     Confederation
    -0.64
     Feld
    -0.64
     Countries
    -0.64
     Ell
    -0.64
     Bohem
    -0.63
    POSITIVE LOGITS
    issue
    0.99
    do
    0.97
    dos
    0.96
    say
    0.96
    upon
    0.95
    did
    0.94
    eat
    0.92
    answer
    0.90
    needed
    0.90
    be
    0.89
    Activations Density 0.054%

    No Known Activations