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. 11-RES-JB
    6. 15850
    Prev
    Next
    INDEX
    Explanations

    terms related to the concept of humanity or human civilization

    oai_token-act-pair · gpt-3.5-turbo

    references to humanity and human existence

    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
     Gillespie
    -0.73
    road
    -0.71
    UFF
    -0.67
    roller
    -0.67
     Rican
    -0.66
     Buckingham
    -0.66
    raph
    -0.66
     Comprehensive
    -0.65
    href
    -0.65
     Angle
    -0.64
    POSITIVE LOGITS
    ankind
    1.22
     beings
    1.20
     inhab
    1.04
     Homo
    1.03
     evolving
    0.98
     humans
    0.96
     humankind
    0.96
     endowed
    0.95
     Humans
    0.93
     enslaved
    0.93
    Activations Density 0.103%

    No Known Activations