© 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. 18848
    Prev
    Next
    INDEX
    Explanations

    proper nouns that appear to be names of people or places

    oai_token-act-pair · gpt-3.5-turbo

    instances of the acronym "AM" and variations related to it

    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
     quit
    -0.70
     dece
    -0.69
     closure
    -0.66
     uncle
    -0.64
     quo
    -0.63
     Fors
    -0.62
     fleeing
    -0.62
     passer
    -0.61
     flee
    -0.61
     voc
    -0.61
    POSITIVE LOGITS
    AM
    4.07
    AMS
    2.91
    am
    2.08
    PM
    1.99
    AMA
    1.86
     AM
    1.81
    AMP
    1.72
    AMI
    1.69
    AME
    1.58
    ams
    1.58
    Activations Density 0.013%

    No Known Activations