Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 5368
    Prev
    Next
    INDEX
    Explanations

    adjectives related to harm or damage

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe various forms of harm or damage inflicted on individuals or communities

    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
     Marketplace
    -0.76
    rosso
    -0.73
     Horizons
    -0.73
    angular
    -0.71
     Kickstarter
    -0.71
     sediment
    -0.67
    Vision
    -0.67
     passport
    -0.65
     Rise
    -0.65
    soDeliveryDate
    -0.64
    POSITIVE LOGITS
     inflicted
    3.02
     inflict
    2.88
     inflicting
    2.74
     inflic
    2.42
     bestowed
    1.22
     perpetrated
    1.22
     wre
    1.13
     heap
    1.06
    licted
    1.06
     wielded
    1.05
    Activations Density 0.054%

    No Known Activations