© 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. 0-RES-JB
    11. 19154
    Prev
    Next
    INDEX
    Explanations

    phrases related to missile or rocket launches

    oai_token-act-pair · gpt-3.5-turbo

    references to different types of launches and bidding processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uana
    -0.73
     situation
    -0.72
     matrix
    -0.72
     disguise
    -0.72
     predicament
    -0.71
     issue
    -0.70
     act
    -0.68
     mistake
    -0.68
     paycheck
    -0.67
     matter
    -0.67
    POSITIVE LOGITS
     launches
    2.60
     bids
    1.89
     dives
    1.64
     cycles
    1.62
     orbits
    1.60
     pitches
    1.59
     deals
    1.58
     recounts
    1.53
     premie
    1.51
     crashes
    1.47
    Activations Density 0.032%

    No Known Activations