© 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. 11-RES-JB
    11. 7818
    Prev
    Next
    INDEX
    Explanations

    phrases indicating importance or significance, such as "the real issue here" or "an important part here."

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the significance or implications of issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    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
    ailability
    -0.65
    CLASSIFIED
    -0.62
    gettable
    -0.61
    dding
    -0.60
     womb
    -0.57
    ustomed
    -0.56
    itted
    -0.56
     ];
    -0.54
    inka
    -0.53
    Done
    -0.53
    POSITIVE LOGITS
     involves
    1.43
     relates
    1.38
     is
    1.35
     boils
    1.34
     revolves
    1.34
     arises
    1.22
     rests
    1.21
     stems
    1.20
     consists
    1.20
     depends
    1.16
    Activations Density 0.252%

    No Known Activations