© 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. 3342
    Prev
    Next
    INDEX
    Explanations

    phrases related to statements or quotes in conversation

    oai_token-act-pair · gpt-3.5-turbo

    negative phrases or sentiments related to inability or unfulfilled desires

    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
    •interesting
    indicates negation

    No Comments

    Negative Logits
    accompan
    -0.65
    Discussion
    -0.61
     Ezek
    -0.58
    VL
    -0.56
    Category
    -0.55
    instead
    -0.55
    rather
    -0.55
    itaire
    -0.55
    renheit
    -0.54
    artney
    -0.54
    POSITIVE LOGITS
     anymore
    1.14
     nor
    0.83
    yet
    0.73
    hin
    0.69
     necessarily
    0.68
    ':
    0.65
     \'
    0.61
     shit
    0.61
     :(
    0.60
     bothered
    0.60
    Activations Density 0.335%

    No Known Activations