© 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. 10-RES-JB
    11. 12538
    Prev
    Next
    INDEX
    Explanations

    references to past experiences or nostalgic sentiments

    oai_token-act-pair · gpt-3.5-turbo

    references to nostalgic themes and expressions of longing for the past

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bsp
    -0.62
    ItemTracker
    -0.60
    Ru
    -0.58
    malink
    -0.58
    Rot
    -0.58
    ctor
    -0.58
     parcel
    -0.58
    redo
    -0.57
    Package
    -0.56
    acts
    -0.56
    POSITIVE LOGITS
     eras
    0.90
     hey
    0.89
     days
    0.88
     era
    0.87
     nostalg
    0.85
     Atari
    0.76
    abouts
    0.75
     circa
    0.71
    adays
    0.71
     Era
    0.69
    Activations Density 0.194%

    No Known Activations