© 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. 7805
    Prev
    Next
    INDEX
    Explanations

    general references or statements

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating generalizations or common sentiments

    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
    letes
    -0.71
    Fax
    -0.71
    fif
    -0.67
    Rub
    -0.65
    Ship
    -0.63
    vana
    -0.63
    Secret
    -0.62
    poke
    -0.61
    Mount
    -0.61
     Shooter
    -0.60
    POSITIVE LOGITS
     relativity
    1.15
     terms
    1.05
     speaking
    1.00
    izations
    0.89
    izing
    0.85
    isations
    0.80
     vicinity
    0.80
    terms
    0.79
    ization
    0.77
    arily
    0.73
    Activations Density 0.044%

    No Known Activations