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

    Neuronpedia

    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. 10-RES-JB
    11. 10530
    Prev
    Next
    INDEX
    Explanations

    locations or places in various regions or cities

    oai_token-act-pair · gpt-3.5-turbo

    references to geographical locations and their surrounding features

    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
    thood
    -0.78
    upon
    -0.73
    ality
    -0.72
    nuts
    -0.71
    mith
    -0.70
    aryl
    -0.69
    uria
    -0.69
    eful
    -0.68
    pell
    -0.67
    boxing
    -0.67
    POSITIVE LOGITS
     outskirts
    1.56
     shores
    1.33
     verge
    1.29
     edge
    1.24
     brink
    1.24
     periphery
    1.20
     horizon
    1.17
     peninsula
    1.14
     coasts
    1.14
     coast
    1.12
    Activations Density 0.120%

    No Known Activations