© 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. 12-RES-JB
    11. 4389
    Prev
    Next
    INDEX
    Explanations

    words related to recounting events or actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    verbs in various forms, particularly 'take' and 'took'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¦ļéĨĴ
    -0.85
    luaj
    -0.75
    mini
    -0.73
    ibel
    -0.73
    enegger
    -0.72
    ãĤ¦ãĤ¹
    -0.71
    uterte
    -0.71
    ionic
    -0.71
    ocaly
    -0.71
     guiActiveUn
    -0.71
    POSITIVE LOGITS
     awhile
    1.19
     precedence
    1.10
     place
    1.06
     longer
    1.02
     months
    0.96
     years
    0.96
     forever
    0.94
     guts
    0.92
     courage
    0.91
     decades
    0.90
    Activations Density 0.062%

    No Known Activations