© 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. 1-RES-JB
    11. 7696
    Prev
    Next
    INDEX
    Explanations

    mentions of the country Cuba

    oai_token-act-pair · gpt-3.5-turbo

    references to Cuba

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yll
    -0.87
    ãĥł
    -0.79
    rawdownloadcloneembedreportprint
    -0.67
    Ü
    -0.67
    ãĥĥãĥĪ
    -0.66
    sign
    -0.64
    ieth
    -0.64
    à
    -0.64
    andr
    -0.64
    âķIJâķIJ
    -0.63
    POSITIVE LOGITS
     Cuba
    1.06
     cigars
    0.93
     Rebels
    0.85
     Libre
    0.82
    illo
    0.79
     Castro
    0.79
    ricular
    0.76
     Missile
    0.75
     Contra
    0.75
    ascus
    0.73
    Activations Density 0.005%

    No Known Activations