© 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. 4-RES-JB
    11. 1700
    Prev
    Next
    INDEX
    Explanations

    highlighted content or emphasis markers such as asterisks

    oai_token-act-pair · gpt-3.5-turbo

    symbols or characters often used to denote emphasis or separation in text

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

    No Comments

    Negative Logits
    uces
    -0.78
     exting
    -0.77
     srf
    -0.76
     curv
    -0.75
    etheless
    -0.74
     foremost
    -0.73
     conservation
    -0.73
     displacement
    -0.72
     radi
    -0.71
     sacrific
    -0.70
    POSITIVE LOGITS
    ***
    0.97
    HAEL
    0.87
    **
    0.86
     ***
    0.80
    NEW
    0.80
    edited
    0.78
    TOP
    0.78
    !/
    0.77
    ERY
    0.77
    Bonus
    0.76
    Activations Density 0.011%

    No Known Activations