© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6669
    Prev
    Next
    INDEX
    Explanations

    the word "sigh" and variations of it

    oai_token-act-pair · gpt-3.5-turbo

    expressions of exasperation or defeat

    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
    atal
    -0.69
    arat
    -0.66
    wana
    -0.64
     constituted
    -0.64
     inclusive
    -0.62
     designated
    -0.62
    osa
    -0.61
    anded
    -0.60
    Operation
    -0.60
     protected
    -0.60
    POSITIVE LOGITS
     sigh
    3.74
     shrug
    2.56
     sighed
    1.89
     frown
    1.84
     yawn
    1.73
     chuckle
    1.71
     shudder
    1.64
     gasp
    1.57
     shrugged
    1.53
     moan
    1.47
    Activations Density 0.022%

    No Known Activations