© 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. 10-RES-JB
    11. 17475
    Prev
    Next
    INDEX
    Explanations

    words related to the state or condition of something, particularly focusing on negative aspects

    oai_token-act-pair · gpt-3.5-turbo

    prepositions indicating states or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     incent
    -0.71
     Norn
    -0.70
    cript
    -0.70
     congratulated
    -0.68
     WATCHED
    -0.67
     listened
    -0.67
     applies
    -0.65
     Introdu
    -0.65
    PLIED
    -0.64
     sentenced
    -0.63
    POSITIVE LOGITS
     turmoil
    1.40
     meltdown
    1.17
     tumult
    1.12
     upheaval
    1.11
     flux
    1.10
     jeopardy
    1.10
     flames
    1.06
     decline
    1.06
     chaos
    1.03
     crisis
    1.03
    Activations Density 0.194%

    No Known Activations