© 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 EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 12625
    Prev
    Next
    INDEX
    Explanations

    proper nouns referring to geographical locations

    oai_token-act-pair · gpt-3.5-turbo

    geographical locations and national identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    govtrack
    -0.75
    loads
    -0.70
    enance
    -0.69
    umers
    -0.66
    */(
    -0.65
    verbs
    -0.64
    ologists
    -0.61
    usions
    -0.61
    ints
    -0.60
    riel
    -0.60
    POSITIVE LOGITS
     era
    0.95
     universe
    0.80
     arena
    0.80
     empire
    0.78
     enclave
    0.77
     realm
    0.76
     National
    0.75
     throne
    0.73
     Museum
    0.73
     landscape
    0.72
    Activations Density 0.526%

    No Known Activations