© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 18311
    Prev
    Next
    INDEX
    Explanations

    phrases with the word "subpoenaed" or similar variations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "subpoenaed."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     spotting
    -0.70
     markup
    -0.63
     Zap
    -0.63
    heads
    -0.62
     sampling
    -0.61
     breaks
    -0.60
     stripes
    -0.60
    packing
    -0.60
     Mash
    -0.59
     Slack
    -0.59
    POSITIVE LOGITS
    aed
    4.53
    ae
    1.40
    æ
    1.25
    aeda
    1.19
     paed
    1.02
    20439
    0.98
    ael
    0.98
    AE
    0.96
    aid
    0.93
    ced
    0.92
    Activations Density 0.016%

    No Known Activations