© 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. 4215
    Prev
    Next
    INDEX
    Explanations

    references to specific political or cultural locations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of notable locations and civic-related references

    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
    pret
    -0.73
    DI
    -0.73
     craft
    -0.71
     paddle
    -0.67
    craft
    -0.66
     microbi
    -0.65
     glob
    -0.65
     modelling
    -0.65
     Targ
    -0.65
     Icelandic
    -0.65
    POSITIVE LOGITS
     Coliseum
    3.87
    iseum
    1.74
     Civic
    1.61
     Chavez
    1.50
     Snapdragon
    1.05
     Contra
    1.01
     Alpine
    0.97
    grounds
    0.96
     Auditor
    0.94
    parable
    0.93
    Activations Density 0.053%

    No Known Activations