© 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. 10-RES-JB
    11. 12249
    Prev
    Next
    INDEX
    Explanations

    abbreviations for locations, especially in the Washington, D.C. area

    oai_token-act-pair · gpt-3.5-turbo

    punctuation, specifically periods marking the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lihood
    -0.64
     fung
    -0.63
    velength
    -0.63
     facult
    -0.60
     chopping
    -0.60
     favour
    -0.58
    bour
    -0.58
    ngth
    -0.58
     underpin
    -0.57
     skelet
    -0.57
    POSITIVE LOGITS
    Va
    1.28
    C
    1.00
    rex
    0.81
    Washington
    0.78
    VA
    0.77
    DC
    0.76
    A
    0.76
    WA
    0.76
    Gray
    0.74
    U
    0.73
    Activations Density 0.016%

    No Known Activations