© 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. 0-RES-JB
    11. 16897
    Prev
    Next
    INDEX
    Explanations

    words related to human skin

    oai_token-act-pair · gpt-3.5-turbo

    references to skin

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    audi
    -0.73
    âĺħâĺħ
    -0.73
    bound
    -0.70
    inqu
    -0.70
    ante
    -0.69
    iott
    -0.69
     Recomm
    -0.68
    trak
    -0.67
    abo
    -0.67
    pend
    -0.66
    POSITIVE LOGITS
     skin
    3.68
    Skin
    2.74
     Skin
    2.72
    skin
    2.70
     skins
    2.59
    skinned
    1.87
     dermat
    1.64
     complexion
    1.62
     scalp
    1.56
     flesh
    1.53
    Activations Density 0.019%

    No Known Activations