© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 22483
    Prev
    Next
    INDEX
    Explanations

    mentions of the country China

    oai_token-act-pair · gpt-3.5-turbo

    country names and their variations

    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
     Rab
    -0.74
     Slate
    -0.72
    orp
    -0.71
    boss
    -0.69
    obar
    -0.68
    ra
    -0.66
    arak
    -0.66
     Floor
    -0.65
    urned
    -0.65
    erb
    -0.64
    POSITIVE LOGITS
    INA
    2.98
    IAN
    2.84
    ISM
    1.40
    ussia
    1.28
    ICS
    1.26
    VILLE
    1.17
    IDA
    1.15
    ENA
    1.10
    SHIP
    1.09
    STON
    1.05
    Activations Density 0.044%

    No Known Activations