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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9-RES-JB
    11. 18417
    Prev
    Next
    INDEX
    Explanations

    sentences containing the word "not" followed by a description or explanation

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize negation or refutation of various concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Oath
    -0.66
    çīĪ
    -0.66
    nect
    -0.62
    imei
    -0.60
    age
    -0.59
    kees
    -0.59
    interstitial
    -0.59
    eger
    -0.59
     Dynasty
    -0.58
    Ult
    -0.56
    POSITIVE LOGITS
     necessarily
    1.23
     merely
    0.97
    withstanding
    0.87
    icably
    0.83
     malice
    0.75
     solely
    0.73
     simply
    0.73
     mere
    0.71
    icable
    0.70
     overtly
    0.69
    Activations Density 0.155%

    No Known Activations