© 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. 0-RES-JB
    11. 16929
    Prev
    Next
    INDEX
    Explanations

    mentions of the city "Sevastopol"

    oai_token-act-pair · gpt-3.5-turbo

    geographical references related to locations and administrative terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Pyth
    -0.71
     AST
    -0.71
    UGE
    -0.68
    animate
    -0.65
     animate
    -0.63
    BE
    -0.61
    DOM
    -0.60
     Singer
    -0.60
     Pillar
    -0.60
     Haitian
    -0.60
    POSITIVE LOGITS
    opol
    4.65
    anton
    1.52
    opoly
    1.29
    opolis
    1.21
    obin
    1.04
    olid
    1.03
    obal
    1.03
    obyl
    1.00
    greg
    1.00
    odon
    0.97
    Activations Density 0.023%

    No Known Activations