© 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. 20860
    Prev
    Next
    INDEX
    Explanations

    words related to moral principles, ethics, and judgement

    oai_token-act-pair · gpt-3.5-turbo

    terms related to sanctity and legality

    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
     Kingdoms
    -0.71
    UNCH
    -0.70
    beit
    -0.68
     Viking
    -0.64
     curfew
    -0.60
    GS
    -0.60
     Sandwich
    -0.60
     Strikes
    -0.60
     tongue
    -0.59
    SPA
    -0.59
    POSITIVE LOGITS
    imon
    1.21
    ified
    1.16
    ification
    1.07
    uitous
    1.05
    ifying
    1.03
    imony
    1.01
    itary
    1.00
    ific
    1.00
    eness
    0.98
    rets
    0.93
    Activations Density 0.016%

    No Known Activations