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. 0-RES-JB
    6. 3084
    Prev
    Next
    INDEX
    Explanations

    mention or focus on someone being skilled or talented

    oai_token-act-pair · gpt-3.5-turbo

    descriptors of talent and skill, particularly related to being "gifted."

    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
    iet
    -0.67
    enter
    -0.66
    ø
    -0.65
    watch
    -0.63
     Date
    -0.61
     Proxy
    -0.61
    alph
    -0.61
     plank
    -0.60
     Norm
    -0.60
     Camel
    -0.60
    POSITIVE LOGITS
     gifted
    3.85
     talented
    1.52
     endowed
    1.48
    ifted
    1.47
     proficient
    1.19
     blessed
    1.16
     entrusted
    1.12
     skilled
    1.11
     disadvantaged
    1.06
     bestowed
    1.06
    Activations Density 0.018%

    No Known Activations