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. 15440
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names, titles, and information related to political events or scandals

    oai_token-act-pair · gpt-3.5-turbo

    references to political figures and their activities

    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
    crim
    -0.77
     ILCS
    -0.75
    Surv
    -0.69
     reven
    -0.69
    usterity
    -0.67
     masc
    -0.67
     amput
    -0.67
     aven
    -0.67
    agnar
    -0.66
    juries
    -0.66
    POSITIVE LOGITS
     memos
    1.28
     Kislyak
    1.26
     memo
    1.17
     briefings
    1.16
     meetings
    1.15
     contacts
    1.10
     contempor
    1.09
     aides
    1.08
     conversations
    1.08
     confidential
    1.05
    Activations Density 0.640%

    No Known Activations