© 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. 8-RES-JB
    11. 17704
    Prev
    Next
    INDEX
    Explanations

    words related to proper names, especially with emphasis on 'ero,' 'ed,' and 'ad.'

    oai_token-act-pair · gpt-3.5-turbo

    specific identifiers and roles related to individuals or entities in a competitive context

    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
    yip
    -0.88
    uala
    -0.71
     dime
    -0.68
    sth
    -0.64
     Mania
    -0.63
    ibrary
    -0.62
    clud
    -0.61
    anguage
    -0.61
     Soda
    -0.59
     Moff
    -0.58
    POSITIVE LOGITS
    enment
    1.05
    bsite
    0.82
    vable
    0.81
    furt
    0.76
    lish
    0.74
    rative
    0.73
    rill
    0.72
    arrass
    0.72
    Allah
    0.71
    ener
    0.69
    Activations Density 0.176%

    No Known Activations