© 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. 11057
    Prev
    Next
    INDEX
    Explanations

    violent actions carried out against individuals

    oai_token-act-pair · gpt-3.5-turbo

    actions related to physical violence and assault

    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
    soType
    -0.87
    Story
    -0.77
    izen
    -0.76
    soDeliveryDate
    -0.74
    Balt
    -0.73
    isphere
    -0.73
    MSN
    -0.72
    natureconservancy
    -0.71
    yet
    -0.70
    stellar
    -0.70
    POSITIVE LOGITS
     bystanders
    0.88
     unarmed
    0.81
     abusive
    0.80
    roc
    0.75
     Stab
    0.74
     violently
    0.73
     unsuccessfully
    0.72
     SWAT
    0.70
     takedown
    0.69
     homophobic
    0.69
    Activations Density 0.132%

    No Known Activations