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. 7418
    Prev
    Next
    INDEX
    Explanations

    names related to the term "Gosnell."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific individuals or entities, particularly names associated with significant public figures or schools

    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
     indu
    -0.71
     Awakening
    -0.71
     Initiative
    -0.70
    Sab
    -0.70
    gdala
    -0.69
     susceptible
    -0.69
     prism
    -0.67
     brace
    -0.67
     Rebellion
    -0.66
     Crusade
    -0.66
    POSITIVE LOGITS
     Gos
    2.78
    ör
    2.05
    Ñĥ
    1.59
    orth
    1.18
     Roe
    1.17
    yth
    1.13
     Loch
    1.05
     Obj
    1.02
     Ãĸ
    0.96
     Poo
    0.95
    Activations Density 0.042%

    No Known Activations