© 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. 11-RES-JB
    11. 7595
    Prev
    Next
    INDEX
    Explanations

    dates mentioned in a format like "Week X" or specific months and years

    oai_token-act-pair · gpt-3.5-turbo

    references to specific weeks and their associated activities or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    âĢ¢âĢ¢
    -0.65
     Masquerade
    -0.62
     transact
    -0.61
     chloride
    -0.60
     liberalism
    -0.60
    agles
    -0.59
     suspic
    -0.59
     Persia
    -0.59
    blers
    -0.59
     Strait
    -0.58
    POSITIVE LOGITS
    days
    1.67
    night
    1.42
    ends
    1.28
    lies
    1.26
    end
    1.11
    DAY
    1.09
    END
    1.03
    day
    1.01
    long
    0.99
    ender
    0.92
    Activations Density 0.102%

    No Known Activations