© 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. 6999
    Prev
    Next
    INDEX
    Explanations

    terms related to legal matters or rules and regulations

    oai_token-act-pair · gpt-3.5-turbo

    statistical references or terms related to populations or demographics

    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
     scrut
    -0.52
    anwhile
    -0.51
     Vaugh
    -0.50
     elig
    -0.50
    ogether
    -0.50
     challeng
    -0.48
    bledon
    -0.47
    gerald
    -0.46
     Downs
    -0.46
    ãĥ¼ãĥĨ
    -0.44
    POSITIVE LOGITS
    É
    0.61
     âĺ
    0.59
     ðŁij
    0.58
     ðŁ
    0.57
     âľ
    0.55
     âľĵ
    0.53
    âĢ
    0.53
     âĢ
    0.53
     (@
    0.53
    xia
    0.53
    Activations Density 3.127%

    No Known Activations