© 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. 8-RES-JB
    11. 15149
    Prev
    Next
    INDEX
    Explanations

    terms related to geographical locations, specifically focusing on Northern areas

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Northern" in relation to geographical or political contexts

    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
    inventoryQuantity
    -0.82
    enger
    -0.80
    awaru
    -0.80
    ifice
    -0.80
    uable
    -0.78
    utenberg
    -0.77
    ulative
    -0.76
    inger
    -0.76
    verb
    -0.76
    earance
    -0.75
    POSITIVE LOGITS
     Hemisphere
    1.30
     hemisphere
    1.14
     Lights
    1.09
     Liberties
    1.00
     Ireland
    0.99
     Territory
    0.97
     Caucasus
    0.89
     Californ
    0.81
    most
    0.79
     Territories
    0.79
    Activations Density 0.027%

    No Known Activations