© 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. 8-RES-JB
    11. 8337
    Prev
    Next
    INDEX
    Explanations

    actions or decisions that are being judged morally or legally, looking for instances of right or wrong behavior

    oai_token-act-pair · gpt-3.5-turbo

    expressions of judgment or assessment of actions

    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
    izen
    -0.76
    cry
    -0.70
    tions
    -0.70
    stories
    -0.70
    hedral
    -0.69
    pop
    -0.68
     Units
    -0.67
     Nightmares
    -0.67
    catentry
    -0.66
    anon
    -0.66
    POSITIVE LOGITS
     negligent
    1.27
     careless
    1.25
     mishand
    1.22
     omission
    1.18
     negligence
    1.16
     neglig
    1.07
     knowingly
    0.99
     willfully
    0.94
     deliberately
    0.93
     intentionally
    0.92
    Activations Density 0.885%

    No Known Activations