© 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. 1161
    Prev
    Next
    INDEX
    Explanations

    words related to embarrassing situations or events

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "embarrassing" and related expressions of embarrassment

    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
    gran
    -0.71
     Reson
    -0.69
    brid
    -0.67
     Tru
    -0.66
    tones
    -0.66
    Creat
    -0.65
    ribution
    -0.64
    tower
    -0.64
    hips
    -0.63
     varies
    -0.63
    POSITIVE LOGITS
     embarrassing
    3.09
     embarrassment
    2.85
     embarrassed
    2.60
     embarrass
    2.40
     humiliating
    2.27
     humiliation
    1.97
     humiliated
    1.95
     humili
    1.76
     shameful
    1.69
     ashamed
    1.63
    Activations Density 0.032%

    No Known Activations