© 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. 3698
    Prev
    Next
    INDEX
    Explanations

    words ending in '-ted'

    oai_token-act-pair · gpt-3.5-turbo

    past tense verbs

    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
     Gutenberg
    -0.80
    vernment
    -0.73
     sqor
    -0.64
     cannabin
    -0.63
     preparations
    -0.62
    soDeliveryDate
    -0.60
     largeDownload
    -0.59
    CRIP
    -0.58
     actionGroup
    -0.58
     evaluating
    -0.58
    POSITIVE LOGITS
    ted
    1.38
    ting
    1.25
    geon
    0.98
    mouth
    0.98
    own
    0.92
    bed
    0.92
    ton
    0.91
    kered
    0.87
    pent
    0.85
    rive
    0.85
    Activations Density 0.008%

    No Known Activations