© 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. 3-RES-JB
    11. 20755
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or events in the past

    oai_token-act-pair · gpt-3.5-turbo

    repeated forms of the word "were"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ocracy
    -0.67
    oire
    -0.66
    ossom
    -0.64
    otic
    -0.64
    lag
    -0.64
    urry
    -0.63
     Matter
    -0.61
    izes
    -0.61
     Raise
    -0.60
     Solution
    -0.59
    POSITIVE LOGITS
    wolves
    1.40
    wolf
    1.23
    nt
    0.97
     supposed
    0.92
     able
    0.91
    hers
    0.82
     greeted
    0.82
     subjected
    0.81
     instrumental
    0.80
     originally
    0.80
    Activations Density 0.179%

    No Known Activations