© 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. 1-RES-JB
    11. 12874
    Prev
    Next
    INDEX
    Explanations

    phrases related to delays or obstructions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "stalled" and related terms indicating delays or interruptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    arers
    -0.85
    obby
    -0.82
    bern
    -0.77
    ophysical
    -0.73
    Reviewer
    -0.72
    ological
    -0.72
    arning
    -0.71
    obbies
    -0.71
    opers
    -0.70
    bors
    -0.69
    POSITIVE LOGITS
     stalled
    1.34
     stall
    1.05
     derail
    0.84
     cooker
    0.82
     slowing
    0.77
     slowdown
    0.77
     slows
    0.75
     burner
    0.73
     retard
    0.73
     stalls
    0.73
    Activations Density 0.007%

    No Known Activations