© 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
    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. 10104
    Prev
    Next
    INDEX
    Explanations

    phrases related to being offended or causing offense

    oai_token-act-pair · gpt-3.5-turbo

    references to being offended or causing offense

    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
    issue
    -0.74
     liner
    -0.72
    runner
    -0.71
    tom
    -0.69
    aver
    -0.69
    negie
    -0.68
    hart
    -0.66
    nosis
    -0.66
    ynthesis
    -0.65
    gravity
    -0.65
    POSITIVE LOGITS
     offended
    0.97
     offend
    0.94
     insulted
    0.84
     indecent
    0.80
    Sax
    0.75
     sexist
    0.74
     Yiannopoulos
    0.73
    ingly
    0.72
     bystanders
    0.72
     offending
    0.71
    Activations Density 0.018%

    No Known Activations