© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 18373
    Prev
    Next
    INDEX
    Explanations

    words related to negative characteristics or situations

    oai_token-act-pair · gpt-3.5-turbo

    the term "nasty" and related negative descriptors

    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
    inoa
    -0.88
    ablished
    -0.81
    inez
    -0.80
    ĸļ
    -0.78
    ibel
    -0.76
    aver
    -0.76
    ingham
    -0.75
    HCR
    -0.74
    authorized
    -0.74
    ucket
    -0.73
    POSITIVE LOGITS
     nasty
    1.05
     surprises
    0.91
     adolesc
    0.79
    soever
    0.78
     poison
    0.76
    bags
    0.76
     venom
    0.76
     spoil
    0.76
     beasts
    0.76
     havoc
    0.75
    Activations Density 0.010%

    No Known Activations