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

    Neuronpedia

    Jacobian LensNEW
    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. 2001
    Prev
    Next
    INDEX
    Explanations

    negative statements or contradictions

    oai_token-act-pair · gpt-3.5-turbo

    negative phrases and rejections of concepts or narratives

    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
    itas
    -0.76
    kees
    -0.74
    Fs
    -0.73
    unders
    -0.68
    units
    -0.67
    })
    -0.67
    ãĤ¼ãĤ¦ãĤ¹
    -0.66
    WAY
    -0.66
    çļ
    -0.65
    anders
    -0.65
    POSITIVE LOGITS
     necessarily
    1.29
     merely
    1.08
     mere
    0.93
    withstanding
    0.91
     meant
    0.84
     flashy
    0.83
     uncommon
    0.79
     solely
    0.79
     rocket
    0.79
     kidding
    0.77
    Activations Density 0.170%

    No Known Activations