© 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. 8-RES-JB
    11. 7643
    Prev
    Next
    INDEX
    Explanations

    profanities and insults

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong frustration or anger

    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
    Not in Any Lists

    No Comments

    Negative Logits
     concess
    -0.77
     preliminary
    -0.75
    raq
    -0.75
     authorization
    -0.75
     migr
    -0.75
     correspond
    -0.74
     enrol
    -0.73
    ön
    -0.73
    lique
    -0.72
    rium
    -0.72
    POSITIVE LOGITS
     Seriously
    1.37
     Fuck
    1.26
     FUCK
    1.22
    Seriously
    1.20
     Especially
    1.17
    Fuck
    1.16
     Anyway
    1.16
     Honestly
    1.14
     Sorry
    1.14
     Stupid
    1.12
    Activations Density 0.698%

    No Known Activations