© 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. 16800
    Prev
    Next
    INDEX
    Explanations

    programming language syntax and commands

    oai_token-act-pair · gpt-3.5-turbo

    technical terms and structured data related to coding or programming contexts

    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
    uga
    -0.71
    agna
    -0.70
    etter
    -0.67
     panc
    -0.64
    Ĥ¬
    -0.63
     Rivera
    -0.61
     prevailed
    -0.61
    ichick
    -0.60
     mosqu
    -0.59
     graz
    -0.59
    POSITIVE LOGITS
     Cheap
    0.73
    Cooldown
    0.70
     Travels
    0.67
     Repeat
    0.66
    ãĤ¦ãĤ¹
    0.65
    200000
    0.65
    arta
    0.64
    arations
    0.64
     sqor
    0.63
    \<
    0.62
    Activations Density 0.727%

    No Known Activations