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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 11-RES-JB
    6. 10684
    Prev
    Next
    INDEX
    Explanations

    words related to disasters, destruction, and crises

    oai_token-act-pair · gpt-3.5-turbo

    terms related to destruction and violence in crises and humanitarian disasters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    theless
    -0.85
    REL
    -0.69
     halves
    -0.67
    itionally
    -0.65
     Honour
    -0.63
     wills
    -0.62
     LIMITED
    -0.60
    rals
    -0.59
    IER
    -0.59
     Compatibility
    -0.58
    POSITIVE LOGITS
     plag
    1.49
     unfolding
    1.21
     wrought
    1.18
     perpetrated
    1.18
     inflicted
    1.16
     occurring
    1.14
     engulf
    1.11
     emanating
    1.08
     happening
    1.06
     endured
    1.02
    Activations Density 0.293%

    No Known Activations