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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1-RES-JB
    11. 18752
    Prev
    Next
    INDEX
    Explanations

    words related to criminal activities and charges

    oai_token-act-pair · gpt-3.5-turbo

    words related to illegal activities, specifically those involving 'imp' which may pertain to a prefix or abbreviation for 'impersonation' or 'improper behavior'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     CU
    -0.74
     Aval
    -0.70
    lees
    -0.68
     Hath
    -0.67
     CLS
    -0.67
     McMaster
    -0.65
    MORE
    -0.65
     Roses
    -0.63
     Kern
    -0.63
    nda
    -0.63
    POSITIVE LOGITS
    imp
    1.30
    ossible
    1.11
    regn
    1.06
    ocalypse
    1.05
    icult
    1.03
    ossibly
    1.00
    ressive
    0.99
    ulse
    0.98
    uppet
    0.97
    aired
    0.96
    Activations Density 0.005%

    No Known Activations