© 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. 11-RES-JB
    11. 18827
    Prev
    Next
    INDEX
    Explanations

    mentions of Earth or specific locations on Earth in a negative context

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a comparison of nations, particularly focusing on status and characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    soDeliveryDate
    -0.84
    enaries
    -0.70
    Äĩ
    -0.63
    zes
    -0.63
     defeats
    -0.62
    raped
    -0.62
    Numbers
    -0.61
    Enabled
    -0.60
    acies
    -0.60
     Cosponsors
    -0.60
    POSITIVE LOGITS
     earth
    1.69
     Earth
    1.49
     planet
    1.45
    Earth
    1.16
     Planet
    1.07
     record
    0.97
    earth
    0.96
     EAR
    0.86
     humankind
    0.86
     campus
    0.85
    Activations Density 0.102%

    No Known Activations