© 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. 10528
    Prev
    Next
    INDEX
    Explanations

    adjectives describing extreme or intense situations

    oai_token-act-pair · gpt-3.5-turbo

    the term "severe" and related variations indicating extreme intensity or conditions

    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
    tein
    -0.81
    aukee
    -0.77
    xxxxxxxx
    -0.76
    ynthesis
    -0.75
    cellent
    -0.74
    hops
    -0.73
    atu
    -0.73
    rium
    -0.73
     Tycoon
    -0.72
    etts
    -0.71
    POSITIVE LOGITS
     severity
    0.91
     earthqu
    0.90
     punishments
    0.87
     consequences
    0.87
     punishment
    0.85
     conflic
    0.85
     understatement
    0.84
     allergic
    0.84
     repercussions
    0.82
     injury
    0.82
    Activations Density 0.020%

    No Known Activations