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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11693
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "first."

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "first" indicating new beginnings or initial experiences

    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
     Vish
    -0.77
    stairs
    -0.76
    bara
    -0.74
    vest
    -0.68
    abus
    -0.64
    within
    -0.63
    skirts
    -0.63
    holes
    -0.62
    blocks
    -0.62
    renches
    -0.62
    POSITIVE LOGITS
     foray
    1.51
     outing
    1.26
     attempt
    1.03
     stint
    1.00
     birthday
    1.00
     appearance
    0.97
     incarnation
    0.97
     cousin
    0.94
     wife
    0.89
     assignment
    0.89
    Activations Density 0.098%

    No Known Activations