© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 20363
    Prev
    Next
    INDEX
    Explanations

    information related to violent incidents, such as attacks, bombings, and casualties

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to violent events or attacks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     commissions
    -0.82
     subpoena
    -0.76
    folio
    -0.75
     invites
    -0.72
     appoint
    -0.72
     persuasion
    -0.72
     candid
    -0.70
     favor
    -0.69
     fid
    -0.69
     appointing
    -0.69
    POSITIVE LOGITS
     deadliest
    0.98
    Authorities
    0.98
    Viol
    0.92
     Survivors
    0.91
     Deaths
    0.90
    Reuters
    0.90
    Thousands
    0.89
    zens
    0.89
     explosions
    0.87
     Dozens
    0.87
    Activations Density 0.770%

    No Known Activations