© 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. 12133
    Prev
    Next
    INDEX
    Explanations

    terms related to physical aggression or violence

    oai_token-act-pair · gpt-3.5-turbo

    references to types of assault or related violent acts

    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
    ocular
    -0.77
    FORE
    -0.68
    ãĤ©
    -0.67
     CFR
    -0.67
     Solitaire
    -0.66
     snipp
    -0.65
    atom
    -0.65
    printed
    -0.64
    ITNESS
    -0.61
     optimizing
    -0.61
    POSITIVE LOGITS
     assault
    0.90
    uous
    0.86
    quez
    0.83
    ively
    0.81
    uously
    0.80
     assaults
    0.79
    entimes
    0.79
     perpetrated
    0.79
    ments
    0.78
    quit
    0.77
    Activations Density 0.020%

    No Known Activations