© 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
    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. 2411
    Prev
    Next
    INDEX
    Explanations

    comparisons or contrasts between different factors or entities

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases emphasizing prioritization or preference over alternatives

    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
    ãĤ¨
    -0.76
    Constructed
    -0.73
    ahime
    -0.70
    onomy
    -0.67
    ango
    -0.63
    estern
    -0.63
    umbn
    -0.62
     largeDownload
    -0.62
    ateur
    -0.61
    OSP
    -0.61
    POSITIVE LOGITS
     anything
    1.18
     ever
    1.11
     any
    1.02
     anybody
    0.97
     anyone
    0.96
     usual
    0.89
     anywhere
    0.88
     vice
    0.77
    ever
    0.76
     slightest
    0.75
    Activations Density 0.109%

    No Known Activations