© 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. 4608
    Prev
    Next
    INDEX
    Explanations

    phrases related to unsafe or dangerous situations

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "unsaturated."

    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
    ŃĶ
    -0.93
     briefs
    -0.77
     lobe
    -0.70
     Hastings
    -0.65
     Krishna
    -0.63
    EVA
    -0.61
     Pillar
    -0.61
     Woodward
    -0.61
     CASE
    -0.60
     MPEG
    -0.60
    POSITIVE LOGITS
    afe
    1.03
    uns
    1.02
    linger
    1.01
    aturated
    1.00
    atisf
    0.99
    ierrez
    0.93
    ourcing
    0.90
    olicited
    0.89
    ustainable
    0.89
    nyder
    0.88
    Activations Density 0.015%

    No Known Activations