© 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. 1-RES-JB
    11. 9548
    Prev
    Next
    INDEX
    Explanations

    profanity or vulgar terms containing the word "cum"

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "cum" or its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĩ
    -0.81
    hips
    -0.76
    Ĭ±
    -0.74
    #$
    -0.73
    chester
    -0.73
    abouts
    -0.71
     Defenders
    -0.71
    ĭ
    -0.68
    ģĸ
    -0.66
     Sawyer
    -0.65
    POSITIVE LOGITS
    ulative
    1.38
    cum
    1.32
     Cum
    1.21
    ming
    1.09
     cum
    1.03
    ulus
    0.98
    ulates
    0.96
    ulate
    0.93
    ceed
    0.90
    brance
    0.89
    Activations Density 0.016%

    No Known Activations