© 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. 7699
    Prev
    Next
    INDEX
    Explanations

    words related to negative aspects or events

    oai_token-act-pair · gpt-3.5-turbo

    references to "mal" terms, particularly those associated with negative or harmful concepts

    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
     Carbuncle
    -0.72
    loo
    -0.68
    nesday
    -0.68
     Hobby
    -0.65
     Fifty
    -0.64
     Emirates
    -0.64
     Clubs
    -0.64
     Jackets
    -0.63
     Defenders
    -0.63
     Wide
    -0.63
    POSITIVE LOGITS
    ignant
    1.19
    colm
    1.08
    practice
    1.07
    adies
    1.04
    absor
    1.02
    formed
    1.01
    igned
    1.00
    igning
    0.97
    cellaneous
    0.92
    adjust
    0.92
    Activations Density 0.023%

    No Known Activations