© 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. 18262
    Prev
    Next
    INDEX
    Explanations

    phrases indicating surprise or lack of surprise

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a lack of surprise or predictability

    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
    ©¶æ
    -0.95
    xia
    -0.86
    tein
    -0.81
    nan
    -0.79
    ript
    -0.76
    alde
    -0.74
    erton
    -0.73
    heimer
    -0.73
    haps
    -0.72
    claimer
    -0.72
    POSITIVE LOGITS
    LY
    0.89
     Squid
    0.88
     surprises
    0.79
     enough
    0.78
    ly
    0.77
     coincidence
    0.75
     revelations
    0.74
    Berry
    0.72
     silence
    0.69
     surprising
    0.69
    Activations Density 0.015%

    No Known Activations