© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 3-RES-JB
    11. 23841
    Prev
    Next
    INDEX
    Explanations

    mentions of criminal activities or illegal behaviors

    oai_token-act-pair · gpt-3.5-turbo

    negative terms associated with crime and societal issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    events
    -0.70
    assies
    -0.65
    rencies
    -0.63
    urches
    -0.62
    sites
    -0.61
    ernels
    -0.60
     establishments
    -0.60
    urrencies
    -0.59
    lees
    -0.59
    ammers
    -0.58
    POSITIVE LOGITS
     standpoint
    0.68
     apology
    0.60
     spokesperson
    0.59
     perspective
    0.57
     unto
    0.55
    ccording
    0.55
     piece
    0.55
    ogram
    0.55
     staffer
    0.55
    ozy
    0.55
    Activations Density 2.363%

    No Known Activations