© 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. 10-RES-JB
    11. 16532
    Prev
    Next
    INDEX
    Explanations

    instances where a contrast or comparison is made emphasizing a specific aspect

    oai_token-act-pair · gpt-3.5-turbo

    statements asserting a condition or situation

    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
    ventures
    -0.75
    igraph
    -0.73
    luaj
    -0.70
    ses
    -0.67
    execute
    -0.67
    ulla
    -0.65
    estial
    -0.65
    enne
    -0.64
    estones
    -0.64
    assies
    -0.64
    POSITIVE LOGITS
     why
    1.30
     especially
    1.15
     understandable
    1.14
     exacerbated
    1.12
     evidenced
    1.09
     partly
    1.09
     compounded
    1.06
     particularly
    1.03
     evident
    0.99
     reflected
    0.99
    Activations Density 0.125%

    No Known Activations