© 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. 2-RES-JB
    11. 1056
    Prev
    Next
    INDEX
    Explanations

    words related to specific locations or cities, particularly "wich" and "wichita"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific geographical locations or cities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    runner
    -0.68
     delinquent
    -0.67
     deaf
    -0.67
    externalActionCode
    -0.65
     distant
    -0.65
     runaway
    -0.63
     destructive
    -0.62
     waning
    -0.60
     frantic
    -0.60
    activated
    -0.58
    POSITIVE LOGITS
    wich
    1.83
    mere
    1.07
    igan
    0.97
    igans
    0.90
    combe
    0.90
    pton
    0.89
    stad
    0.89
    oche
    0.87
    ridge
    0.86
    bury
    0.85
    Activations Density 0.004%

    No Known Activations