© 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. 4-RES-JB
    11. 22203
    Prev
    Next
    INDEX
    Explanations

    the word "surprise" or similar variations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a lack of surprise or expected outcomes

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

    No Comments

    Negative Logits
    eatures
    -0.84
    minster
    -0.79
    chnology
    -0.77
    İĭ
    -0.76
    rote
    -0.74
    adle
    -0.72
    folios
    -0.71
    nai
    -0.71
    rogram
    -0.70
    ettel
    -0.70
    POSITIVE LOGITS
     whatsoever
    0.87
     anymore
    0.82
     nor
    0.78
     surprises
    0.67
     surprise
    0.66
    imaru
    0.65
     prompts
    0.64
     why
    0.64
     enough
    0.63
    REDACTED
    0.63
    Activations Density 0.021%

    No Known Activations