© 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 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. 7-RES-JB
    11. 1708
    Prev
    Next
    INDEX
    Explanations

    descriptions of human physical actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of punctuation that separate actions or descriptions in a narrative context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    eus
    -0.72
    UF
    -0.71
    Commerce
    -0.70
    hack
    -0.70
    FP
    -0.68
    ©
    -0.66
    itionally
    -0.66
     Confederation
    -0.66
    ensor
    -0.65
    ãĤ´ãĥ³
    -0.65
    POSITIVE LOGITS
     grinning
    1.08
     clutching
    0.99
     whispering
    0.98
     dangling
    0.92
     nodding
    0.91
     echoing
    0.91
     wiping
    0.90
     flanked
    0.90
     trembling
    0.89
     staring
    0.89
    Activations Density 0.346%

    No Known Activations