© 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. 8-RES-JB
    11. 1366
    Prev
    Next
    INDEX
    Explanations

    phrases indicating surprise or disbelief

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing a sense of denial or lack of awareness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 8 - surprise

    No Comments

    Negative Logits
    idon
    -0.86
    rend
    -0.73
    abre
    -0.72
    ubi
    -0.69
    rog
    -0.69
    ijing
    -0.67
    osity
    -0.67
    orm
    -0.66
    center
    -0.65
    aim
    -0.65
    POSITIVE LOGITS
     remotely
    1.29
     bothered
    0.99
     bother
    0.97
     bothering
    0.95
     pretend
    0.88
     halfway
    0.79
     close
    0.78
     kidding
    0.78
     mention
    0.77
     hint
    0.77
    Activations Density 0.049%

    No Known Activations