© 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. 10-RES-JB
    11. 2910
    Prev
    Next
    INDEX
    Explanations

    the word "does" followed by another word, implying actions being performed or not performed

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "does not" indicating negation or contrast in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    boarding
    -0.80
    estern
    -0.65
     Seasons
    -0.63
    ansas
    -0.62
    isu
    -0.62
     palms
    -0.61
     Learns
    -0.61
     este
    -0.59
     Ready
    -0.58
     Printed
    -0.58
    POSITIVE LOGITS
    vet
    1.17
     NOT
    1.05
    oms
    0.97
     not
    0.97
     nothing
    0.94
     exist
    0.92
    oming
    0.89
     indeed
    0.87
     seem
    0.87
     occur
    0.87
    Activations Density 0.065%

    No Known Activations