© 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. 2234
    Prev
    Next
    INDEX
    Explanations

    adjectives with negative connotations preceded by "much-"

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the presence of significant issues or problems

    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
     Phi
    -0.75
     Compass
    -0.75
     Solitaire
    -0.74
     Creator
    -0.74
     Rocks
    -0.74
     Borders
    -0.73
     SOL
    -0.72
     Polo
    -0.71
     Suit
    -0.70
     ROCK
    -0.68
    POSITIVE LOGITS
    needed
    1.65
    anticipated
    1.46
    awaited
    1.31
    used
    1.27
    disc
    1.24
    earned
    1.23
    request
    1.21
    respected
    1.19
    reported
    1.18
    advert
    1.18
    Activations Density 0.034%

    No Known Activations