© 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. 2-RES-JB
    11. 456
    Prev
    Next
    INDEX
    Explanations

    the word "Song" when it appears as a proper noun

    oai_token-act-pair · gpt-4-turbo

    keywords related to the word "Song"

    oai_token-act-pair · gpt-3.5-turbo

    references to musical pieces or songs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    orate
    -0.83
     Kear
    -0.77
    ardless
    -0.72
     Pradesh
    -0.71
    rontal
    -0.69
    etheless
    -0.69
     LAT
    -0.64
     saline
    -0.61
    alsh
    -0.60
    izabeth
    -0.59
    POSITIVE LOGITS
    Song
    1.27
     Song
    1.25
    song
    1.10
    writers
    1.08
    writer
    1.07
     lyrics
    1.05
    bird
    1.02
     Songs
    1.01
    writing
    1.01
    stress
    0.96
    Activations Density 0.009%

    No Known Activations