© 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. 1-RES-JB
    11. 3504
    Prev
    Next
    INDEX
    Explanations

    the letter "n" followed by a numerical value, possibly used as a code or identifier

    oai_token-act-pair · gpt-3.5-turbo

    recurrent use of the letter 'n'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    awaru
    -0.77
    rero
    -0.70
    endas
    -0.67
    姫
    -0.66
    MODE
    -0.66
     XY
    -0.65
     CLSID
    -0.63
    gerald
    -0.61
    lishing
    -0.60
     Prix
    -0.60
    POSITIVE LOGITS
    imble
    1.11
    igg
    1.10
    usra
    1.03
    aught
    1.00
    ought
    0.99
    atal
    0.98
    umpy
    0.95
    ailing
    0.95
    ectar
    0.93
    uggets
    0.90
    Activations Density 0.018%

    No Known Activations