© 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. 0-RES-JB
    11. 2362
    Prev
    Next
    INDEX
    Explanations

    words related to physical damage or destruction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "damaged" and its various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     conven
    -0.73
    Introdu
    -0.70
    ocracy
    -0.70
    Auth
    -0.69
     execution
    -0.67
     Integration
    -0.65
    ream
    -0.65
    Exec
    -0.65
    Trend
    -0.64
    Stra
    -0.63
    POSITIVE LOGITS
     damaged
    3.47
     destroyed
    1.98
     wrecked
    1.98
     harmed
    1.84
     injured
    1.79
     crippled
    1.73
     ruined
    1.71
     defective
    1.71
     wounded
    1.71
     repaired
    1.70
    Activations Density 0.018%

    No Known Activations