© 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. 1-RES-JB
    11. 15979
    Prev
    Next
    INDEX
    Explanations

    intense and negative adjectives used to describe events or situations

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of extreme violence or distressing events

    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
    ebus
    -0.78
    spring
    -0.75
    pai
    -0.75
    arta
    -0.72
    inion
    -0.70
    berman
    -0.69
    ership
    -0.69
    irk
    -0.69
    hibition
    -0.67
    sama
    -0.66
    POSITIVE LOGITS
     horrors
    0.93
     atrocities
    0.87
     gruesome
    0.85
     earthqu
    0.84
     horrific
    0.82
    geoning
    0.80
     nightmares
    0.76
     corrid
    0.74
     carnage
    0.73
     atroc
    0.73
    Activations Density 0.024%

    No Known Activations