© 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. 10-RES-JB
    11. 2223
    Prev
    Next
    INDEX
    Explanations

    large numerical quantities, particularly in the range of tens and hundreds

    oai_token-act-pair · gpt-3.5-turbo

    phrases that refer to large quantities or amounts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Destination
    -0.60
     BAL
    -0.56
     Mate
    -0.55
     Situation
    -0.55
     Rom
    -0.53
    agate
    -0.53
     DG
    -0.52
     mosqu
    -0.51
     Friends
    -0.51
     Messenger
    -0.51
    POSITIVE LOGITS
    ourcing
    1.11
    ousands
    0.96
    ourced
    0.91
    itudes
    0.89
    entimes
    0.88
    eteenth
    0.87
    ivating
    0.86
    omet
    0.85
    illion
    0.84
    hing
    0.84
    Activations Density 0.024%

    No Known Activations