© 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. 10-RES-JB
    11. 18913
    Prev
    Next
    INDEX
    Explanations

    mentions of world leaders and government actions

    oai_token-act-pair · gpt-3.5-turbo

    the verb "has" and its variations to denote possession or attributes regarding individuals or entities

    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
    Reading
    -0.70
    iscovery
    -0.63
    DX
    -0.62
    wrap
    -0.61
     IMAGES
    -0.61
    native
    -0.60
    ribute
    -0.60
     Apart
    -0.59
    Interested
    -0.59
     Pair
    -0.57
    POSITIVE LOGITS
     vowed
    1.36
     been
    1.31
     insisted
    1.20
     repeatedly
    1.19
     pledged
    1.18
     struggled
    1.17
     campaigned
    1.16
     signaled
    1.15
     refused
    1.15
     resisted
    1.14
    Activations Density 0.231%

    No Known Activations