© 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. 6-RES-JB
    11. 2904
    Prev
    Next
    INDEX
    Explanations

    words related to strong negative emotions like horror, shock, and being appalled

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong negative emotions, particularly related to shock or horror

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amins
    -0.71
    impro
    -0.71
    estate
    -0.71
    cially
    -0.70
    aho
    -0.69
    eworks
    -0.69
    ramid
    -0.68
    icipated
    -0.66
    ello
    -0.65
    enture
    -0.65
    POSITIVE LOGITS
     aback
    0.82
    ĸļ
    0.80
    ingly
    0.74
     silence
    0.72
    ãĤ¦ãĤ¹
    0.71
     disbelief
    0.70
     onlook
    0.68
     horrified
    0.68
    urous
    0.66
     bystanders
    0.65
    Activations Density 0.065%

    No Known Activations