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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 7-RES-JB
    11. 14577
    Prev
    Next
    INDEX
    Explanations

    phrases related to potential negative effects or consequences

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate serious negative impacts or consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    abase
    -0.73
     reference
    -0.69
    atography
    -0.69
    itely
    -0.68
    obbies
    -0.68
     alphabet
    -0.67
    mates
    -0.66
    \":
    -0.66
     witty
    -0.66
     honorable
    -0.65
    POSITIVE LOGITS
     havoc
    1.20
     destabil
    1.13
     repercussions
    1.06
     impacts
    1.03
     irreversible
    0.99
     exacerb
    0.98
     jeopard
    0.98
     devastating
    0.98
     adversely
    0.98
     landsl
    0.97
    Activations Density 1.063%

    No Known Activations