© 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. 17101
    Prev
    Next
    INDEX
    Explanations

    phrases indicating emphasis or importance, often related to comparisons

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the degree or extent of something

    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
     STA
    -0.72
    ounded
    -0.69
    ersive
    -0.67
    aleb
    -0.66
     Compliance
    -0.64
    iets
    -0.64
    eele
    -0.63
    hua
    -0.62
    É
    -0.61
     Topics
    -0.61
    POSITIVE LOGITS
     imaginable
    1.26
    except
    0.96
     except
    0.96
     conceivable
    0.91
     nodd
    0.84
    nut
    0.77
     thereafter
    0.73
     sane
    0.72
     whatsoever
    0.72
    Except
    0.70
    Activations Density 0.231%

    No Known Activations