© 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. 3-RES-JB
    11. 13395
    Prev
    Next
    INDEX
    Explanations

    text related to people being openly part of a marginalized group or openly expressing controversial opinions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "openly."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anan
    -0.76
    akeru
    -0.72
     Summoner
    -0.67
    nesota
    -0.66
    error
    -0.65
    enny
    -0.65
    eer
    -0.65
    Guy
    -0.64
    ammy
    -0.64
    oooo
    -0.64
    POSITIVE LOGITS
     openly
    0.81
    odox
    0.80
    lishing
    0.76
     lit
    0.74
     acknowledged
    0.73
     confess
    0.72
     contradicted
    0.72
     disreg
    0.71
     expressed
    0.71
    uously
    0.70
    Activations Density 0.012%

    No Known Activations