© 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. 2-RES-JB
    11. 3211
    Prev
    Next
    INDEX
    Explanations

    the words "I'm" followed by a verb

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "I'm" indicating personal statements or expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Reduce
    -0.64
     separates
    -0.63
     Tactics
    -0.61
     entails
    -0.59
     Stim
    -0.59
     manifests
    -0.58
     occurs
    -0.58
     Cooldown
    -0.58
     mater
    -0.57
     Faces
    -0.57
    POSITIVE LOGITS
     gonna
    1.31
     gotta
    0.94
     going
    0.88
     glad
    0.83
     not
    0.83
     guessing
    0.82
     supposed
    0.82
     hoping
    0.81
     lucky
    0.81
     sorry
    0.78
    Activations Density 0.030%

    No Known Activations