© 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. 16002
    Prev
    Next
    INDEX
    Explanations

    phrases indicating an understatement

    oai_token-act-pair · gpt-3.5-turbo

    phrases that convey degrees of understatement or express opinions about something being described

    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
     coerc
    -0.75
    nels
    -0.71
     interven
    -0.69
    doors
    -0.67
     Procedures
    -0.61
    iste
    -0.60
     harms
    -0.59
    ictive
    -0.58
    hooting
    -0.57
     ens
    -0.57
    POSITIVE LOGITS
     understatement
    1.10
     recalling
    0.88
     hindsight
    0.87
     recall
    0.87
     recol
    0.85
     credit
    0.85
     recollection
    0.84
     reminis
    0.84
     compare
    0.83
     remem
    0.82
    Activations Density 0.434%

    No Known Activations