© 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. 6-RES-JB
    11. 19730
    Prev
    Next
    INDEX
    Explanations

    words related to negative or criminal activities, such as trouble, jail, arrested, or bankrupt

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to legal consequences and punishments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ript
    -0.66
     refresh
    -0.58
    orney
    -0.57
     supplemented
    -0.56
    RFC
    -0.55
    quart
    -0.54
     curated
    -0.54
    visual
    -0.54
     BST
    -0.53
     inaug
    -0.53
    POSITIVE LOGITS
     persecution
    0.69
    ogenic
    0.68
     infring
    0.68
     offenders
    0.65
     martyr
    0.64
     havoc
    0.64
     victims
    0.63
     whistleblowers
    0.63
     criminals
    0.63
     Prevention
    0.62
    Activations Density 0.568%

    No Known Activations