Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 5730
    Prev
    Next
    INDEX
    Explanations

    locations or directions, specifically focusing on directions like "eastern" and "western."

    oai_token-act-pair · gpt-3.5-turbo

    references to geographic directions, specifically "eastern" and "western"

    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
    oku
    -0.90
    vous
    -0.89
    Sharp
    -0.87
    tle
    -0.85
    INAL
    -0.85
    Poll
    -0.83
    Applic
    -0.83
    BALL
    -0.81
     istg
    -0.80
    Deal
    -0.78
    POSITIVE LOGITS
     hemisphere
    1.08
    side
    0.89
    most
    0.85
     regions
    0.85
     bloc
    0.84
     civilization
    0.81
     shores
    0.81
     Aleppo
    0.79
     provinces
    0.78
     region
    0.77
    Activations Density 0.015%

    No Known Activations