© 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. 8-RES-JB
    11. 3112
    Prev
    Next
    INDEX
    Explanations

    references to charts or diagrams

    oai_token-act-pair · gpt-3.5-turbo

    references to charts and data visualizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ignt
    -0.71
     Romeo
    -0.70
    Interstitial
    -0.63
    ntil
    -0.63
     Cic
    -0.59
    INGTON
    -0.58
    IRE
    -0.57
    vae
    -0.57
     Paulo
    -0.56
    iqueness
    -0.56
    POSITIVE LOGITS
    ered
    1.21
    chart
    0.95
    ued
    0.93
    ilion
    0.83
    erer
    0.82
    ographer
    0.80
    eer
    0.80
    erers
    0.80
    rist
    0.79
    makers
    0.78
    Activations Density 0.037%

    No Known Activations