© 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. 8-RES-JB
    11. 3839
    Prev
    Next
    INDEX
    Explanations

    numerical data and measurements

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to numerical data and counts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    the first few tokens after a ,

    Explanation Uploaded by User
    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
    •Skip Features
    its a comma skip feature

    No Comments

    Negative Logits
    otonin
    -0.68
    ptin
    -0.63
    ggle
    -0.63
    igm
    -0.63
    ():
    -0.62
    ogly
    -0.62
    iatus
    -0.61
    otomy
    -0.61
     SAP
    -0.60
    verages
    -0.60
    POSITIVE LOGITS
     notwithstanding
    1.41
     said
    0.86
     included
    0.83
     preferring
    0.79
    nown
    0.75
     argues
    0.75
     says
    0.72
     confirmed
    0.70
     concedes
    0.67
     SAY
    0.67
    Activations Density 0.808%

    No Known Activations