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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10-RES-JB
    11. 16423
    Prev
    Next
    INDEX
    Explanations

    words related to physical harm and injury

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of violence and serious injuries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anic
    -0.74
    ifully
    -0.74
     Citation
    -0.70
    acious
    -0.69
    urous
    -0.68
    yip
    -0.67
    antry
    -0.66
     Mutual
    -0.66
     Parenthood
    -0.66
    iful
    -0.65
    POSITIVE LOGITS
     injured
    1.26
     wounded
    1.17
     injuring
    1.16
     damaged
    1.14
     harmed
    1.10
     impacted
    1.01
     injure
    1.00
     impaired
    1.00
     deficient
    0.98
     affected
    0.98
    Activations Density 0.067%

    No Known Activations