© 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. 4-RES-JB
    11. 20813
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically related to titles or names

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to specific titles or names, particularly in media and gaming contexts

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

    No Comments

    Negative Logits
    lier
    -0.80
    aneously
    -0.78
    activity
    -0.75
     forestry
    -0.74
    istically
    -0.72
     exception
    -0.72
    ren
    -0.69
    ously
    -0.69
     fisherman
    -0.67
    pton
    -0.66
    POSITIVE LOGITS
    hift
    1.18
    peed
    1.15
    aurus
    1.08
    mith
    1.08
     Ago
    1.08
    hops
    1.04
    pread
    1.03
    pace
    1.02
    ystem
    0.99
     Parties
    0.98
    Activations Density 0.172%

    No Known Activations