© 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. 8-RES-JB
    11. 504
    Prev
    Next
    INDEX
    Explanations

    words related to challenges, decisions, opportunities, and requests

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to restrictions and evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •single token firings
    strongly activates on single tokens like " disadvantage", " inconvenience", " objection", " burden" and appears on those more weakly in longer contexts

    No Comments

    Negative Logits
     challeng
    -0.77
    DAQ
    -0.72
     Seym
    -0.69
     Moroc
    -0.69
     withd
    -0.68
     tradem
    -0.65
    enegger
    -0.65
     corrid
    -0.64
     notor
    -0.63
     Noon
    -0.60
    POSITIVE LOGITS
    lessly
    0.74
     ]
    0.70
    ¶
    0.68
     prompt
    0.59
     unfocusedRange
    0.58
    osa
    0.58
    imon
    0.55
     +=
    0.55
    isms
    0.55
     ][
    0.54
    Activations Density 0.293%

    No Known Activations