© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 11-RES-JB
    11. 22068
    Prev
    Next
    INDEX
    Explanations

    quotation marks indicating dialogue

    oai_token-act-pair · gpt-3.5-turbo

    direct quotes or dialogue in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     eleph
    -0.96
     describ
    -0.93
     authorized
    -0.92
     accomp
    -0.92
     subsidized
    -0.89
     censored
    -0.89
     tyrann
    -0.88
     discern
    -0.88
     rul
    -0.87
     exting
    -0.87
    POSITIVE LOGITS
    Obviously
    1.85
    We
    1.77
    I
    1.73
    Hopefully
    1.73
    It
    1.73
    Personally
    1.73
    Being
    1.71
    Everybody
    1.69
    Everyone
    1.67
    Firstly
    1.66
    Activations Density 0.127%

    No Known Activations