© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 19420
    Prev
    Next
    INDEX
    Explanations

    words ending in '-y', either with a vowel before the 'y' or with 'qu' before the 'y'

    oai_token-act-pair · gpt-3.5-turbo

    diminutive or affectionate suffixes and related word forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Polo
    -0.77
    Maps
    -0.71
     Cob
    -0.68
     Marriage
    -0.65
     Siber
    -0.64
     NX
    -0.64
     Xia
    -0.64
    Building
    -0.63
    Weather
    -0.61
     Frie
    -0.61
    POSITIVE LOGITS
    eenth
    0.93
    etheless
    0.85
    anging
    0.77
     suck
    0.73
    terday
    0.73
     mortals
    0.73
    urnal
    0.73
    ither
    0.72
    dozen
    0.71
    actory
    0.71
    Activations Density 0.183%

    No Known Activations