© 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 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. 9-RES-JB
    11. 15829
    Prev
    Next
    INDEX
    Explanations

    metadata and social media mentions, specifically items enclosed in brackets

    oai_token-act-pair · gpt-3.5-turbo

    various types of brackets or symbols often involving lists or updates

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     poisoning
    -0.78
     converted
    -0.74
     equivalents
    -0.74
     appropri
    -0.74
     transported
    -0.71
     continuity
    -0.71
     setting
    -0.71
     virginity
    -0.71
     values
    -0.70
     allowance
    -0.70
    POSITIVE LOGITS
    âĢ¦]
    1.56
    ...]
    1.51
    np
    1.27
    Native
    1.23
    youtube
    1.23
    email
    1.21
    Laughs
    1.20
    !]
    1.19
    UPDATE
    1.19
    Edited
    1.19
    Activations Density 0.025%

    No Known Activations