© 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. 10-RES-JB
    11. 2691
    Prev
    Next
    INDEX
    Explanations

    numerical statistics or margins of victory

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating vote margins and electoral results

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jad
    -0.77
    ociate
    -0.73
    SPONSORED
    -0.71
    OPA
    -0.67
    hai
    -0.67
    igrate
    -0.62
    iHUD
    -0.62
    =(
    -0.62
    aid
    -0.62
    rex
    -0.61
    POSITIVE LOGITS
     landslide
    1.15
     virtue
    1.02
     margins
    0.91
     sheer
    0.90
     unanimous
    0.89
    election
    0.88
     leaps
    0.87
     knockout
    0.83
     landsl
    0.83
     scores
    0.82
    Activations Density 0.071%

    No Known Activations