© 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. 20180
    Prev
    Next
    INDEX
    Explanations

    mentions of thinness or slenderness

    oai_token-act-pair · gpt-3.5-turbo

    references to thinness or thin materials

    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
    ILLE
    -0.79
    USA
    -0.74
    ECH
    -0.74
    UN
    -0.72
    PRES
    -0.71
     USA
    -0.70
    Particip
    -0.69
    andise
    -0.69
     Manual
    -0.69
     Judicial
    -0.68
    POSITIVE LOGITS
     thin
    1.17
    thin
    1.10
     thinner
    1.05
    ning
    0.95
    ned
    0.95
     tiss
    0.94
     thickness
    0.92
    ethy
    0.91
    ners
    0.91
     coats
    0.89
    Activations Density 0.005%

    No Known Activations