© 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. 17270
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences indicating severe damage, injury, or restriction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "severely" indicating serious or intense conditions

    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
    tein
    -0.85
    ynthesis
    -0.79
    akeru
    -0.77
    ioch
    -0.77
    atu
    -0.77
    amide
    -0.76
    ieri
    -0.73
    soDeliveryDate
    -0.73
    eer
    -0.71
    rium
    -0.70
    POSITIVE LOGITS
     punished
    0.92
     inflicted
    0.87
     exting
    0.82
     deteriorated
    0.82
     retarded
    0.82
     cripp
    0.81
     wounding
    0.81
     injuring
    0.80
     harmed
    0.80
     deficient
    0.80
    Activations Density 0.012%

    No Known Activations