© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 6423
    Prev
    Next
    INDEX
    Explanations

    email newsletter content such as daily dispatches and editor picks

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific newsletter or publication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    apesh
    -0.80
    abase
    -0.76
    paren
    -0.73
    nown
    -0.70
    amon
    -0.67
    uthor
    -0.67
    ivas
    -0.66
    inventoryQuantity
    -0.66
    istant
    -0.65
    ethe
    -0.64
    POSITIVE LOGITS
     Picks
    0.94
     Newsletter
    0.72
     Opinion
    0.72
     Brief
    0.68
    Ni
    0.66
     Juice
    0.65
     Recipes
    0.65
     newsletter
    0.63
     Travels
    0.61
     Fever
    0.60
    Activations Density 0.041%

    No Known Activations