© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 530
    Prev
    Next
    INDEX
    Explanations

    words ending in 'un'

    oai_token-act-pair · gpt-3.5-turbo

    the prefix 'un' indicating negation or opposition in various contexts

    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
    Footnote
    -0.66
     mileage
    -0.65
     endpoint
    -0.65
     literacy
    -0.62
     Front
    -0.61
     roadside
    -0.58
     front
    -0.58
     embr
    -0.58
     evidenced
    -0.58
     plaster
    -0.57
    POSITIVE LOGITS
    un
    3.89
    uns
    2.39
    una
    2.01
    uni
    1.96
    UN
    1.89
    une
    1.77
    Un
    1.72
    unc
    1.72
    uno
    1.62
    unin
    1.59
    Activations Density 0.022%

    No Known Activations