© 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. 4-RES-JB
    11. 10627
    Prev
    Next
    INDEX
    Explanations

    short words that seem to be placeholders or errors in the text, possibly arising from encoding issues

    oai_token-act-pair · gpt-3.5-turbo

    negated forms of the verb "to be."

    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
    Reviewer
    -0.86
     SetTextColor
    -0.71
    CVE
    -0.68
     Flavoring
    -0.66
     Gleaming
    -0.63
     Radiation
    -0.63
    Intern
    -0.62
    £ı
    -0.61
     Variant
    -0.60
    soType
    -0.60
    POSITIVE LOGITS
    unes
    1.02
    cha
    0.95
    weet
    0.92
    une
    0.90
    otally
    0.89
    itled
    0.89
    ruck
    0.89
    ucked
    0.89
    aper
    0.88
    actic
    0.87
    Activations Density 0.066%

    No Known Activations