© 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. 0-RES-JB
    11. 18172
    Prev
    Next
    INDEX
    Explanations

    words related to being "busted" or "distrusted"

    oai_token-act-pair · gpt-3.5-turbo

    words related to feelings of being overwhelmed or distressed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Charl
    -0.69
     mono
    -0.65
     Dr
    -0.64
     prism
    -0.63
     Kat
    -0.63
     lip
    -0.63
     HAL
    -0.61
     styl
    -0.61
     LC
    -0.60
     Kenn
    -0.59
    POSITIVE LOGITS
    usted
    4.51
    usting
    3.40
    UST
    1.79
    ust
    1.48
    usters
    1.31
    uster
    1.27
    athered
    1.11
    rust
    1.02
    agged
    1.01
    icked
    1.00
    Activations Density 0.012%

    No Known Activations