© 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. 10693
    Prev
    Next
    INDEX
    Explanations

    names or references related to Persians

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific organization or group of people (Persians)

    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
    payer
    -0.68
    Reviewer
    -0.68
    ITAL
    -0.67
     front
    -0.65
    URN
    -0.64
    REM
    -0.63
    RAFT
    -0.62
    llah
    -0.62
     Aval
    -0.61
    yrinth
    -0.61
    POSITIVE LOGITS
    istence
    1.41
    istent
    1.33
    istant
    1.14
    isting
    1.02
    pect
    0.95
    hing
    0.94
    istance
    0.93
    piration
    0.92
    idium
    0.91
    essor
    0.90
    Activations Density 0.031%

    No Known Activations