© 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. 3-RES-JB
    11. 12564
    Prev
    Next
    INDEX
    Explanations

    references to government agencies

    oai_token-act-pair · gpt-3.5-turbo

    mentions of governmental or organizational agencies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lihood
    -0.77
     Tea
    -0.71
    kai
    -0.70
    aston
    -0.69
     Arn
    -0.69
     Dear
    -0.68
     Shields
    -0.66
    roleum
    -0.66
     Klu
    -0.65
     Plate
    -0.64
    POSITIVE LOGITS
     agency
    0.95
     agencies
    0.92
    encies
    0.83
    acco
    0.81
     tasked
    0.80
     bureaucracy
    0.79
    ional
    0.76
    ENCY
    0.75
     contractor
    0.75
     administrator
    0.74
    Activations Density 0.020%

    No Known Activations