© 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. 15805
    Prev
    Next
    INDEX
    Explanations

    cities or locations based on the pattern "cityname[number or abbreviation]"

    oai_token-act-pair · gpt-3.5-turbo

    references to locations or institutions that include "Amherst."

    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
    advertisement
    -0.75
    glers
    -0.68
    ickr
    -0.63
    Dest
    -0.62
     NK
    -0.61
     Romania
    -0.61
     Manafort
    -0.59
     Myster
    -0.58
     Mao
    -0.58
     Typhoon
    -0.58
    POSITIVE LOGITS
    endment
    0.89
    assador
    0.78
    cade
    0.72
    lehem
    0.69
    ixture
    0.68
    ments
    0.67
    urate
    0.66
    ixtures
    0.65
    ocent
    0.65
    ICAN
    0.65
    Activations Density 0.078%

    No Known Activations