© 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. 1-RES-JB
    11. 16415
    Prev
    Next
    INDEX
    Explanations

    mentions of armed assailants

    oai_token-act-pair · gpt-3.5-turbo

    references to gunmen and violence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ï¸
    -0.78
     Cobb
    -0.76
    âĨij
    -0.74
    geist
    -0.73
    Brew
    -0.73
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    -0.72
    craft
    -0.71
    FH
    -0.69
    Discuss
    -0.69
    mind
    -0.68
    POSITIVE LOGITS
     gunmen
    1.16
    DragonMagazine
    1.10
     assassins
    0.85
     millenn
    0.82
     challeng
    0.82
     accomp
    0.81
    ilitary
    0.78
     slaughtered
    0.75
     slit
    0.73
     contrace
    0.72
    Activations Density 0.017%

    No Known Activations