© 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. 0-RES-JB
    11. 5392
    Prev
    Next
    INDEX
    Explanations

    descriptions related to various tones or qualities of sound or color

    oai_token-act-pair · gpt-3.5-turbo

    references to hues and tones in a visual or musical context

    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
     indisp
    -0.67
     basket
    -0.63
    ezvous
    -0.63
    iken
    -0.63
     clot
    -0.63
     bookmark
    -0.62
    CHA
    -0.61
     Alone
    -0.60
     bottleneck
    -0.60
     margin
    -0.59
    POSITIVE LOGITS
    tones
    4.52
     tones
    2.52
    tone
    2.40
     vocals
    1.35
     frequencies
    1.25
     blues
    1.25
     shades
    1.07
    undy
    1.06
     accents
    1.04
    parts
    1.03
    Activations Density 0.027%

    No Known Activations