© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 12952
    Prev
    Next
    INDEX
    Explanations

    references to the mouth and actions involving the mouth, including both positive and negative contexts

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "mouth" and its variations

    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
    cycles
    -0.72
     Disabled
    -0.70
     Valiant
    -0.67
    rises
    -0.67
    imore
    -0.67
     Rates
    -0.65
    Games
    -0.63
    ifi
    -0.63
    rad
    -0.62
    Data
    -0.61
    POSITIVE LOGITS
     mouth
    3.99
     mouths
    3.16
     Mouth
    2.75
     lips
    2.21
    mouth
    2.21
     tongue
    1.78
     throat
    1.63
     tongues
    1.55
     jaws
    1.52
     jaw
    1.45
    Activations Density 0.007%

    No Known Activations