© 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. 0-RES-JB
    11. 18093
    Prev
    Next
    INDEX
    Explanations

    items listed in a sequence

    oai_token-act-pair · gpt-3.5-turbo

    mentions of lists

    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
    oa
    -0.74
    ammed
    -0.72
     dstg
    -0.67
     srfAttach
    -0.63
    fitting
    -0.63
    Ħ¢
    -0.62
    ooth
    -0.62
    zees
    -0.62
    stead
    -0.62
    lings
    -0.61
    POSITIVE LOGITS
     list
    3.68
     lists
    2.80
     List
    2.51
    list
    2.49
     Lists
    2.28
     LIST
    2.20
    List
    2.13
    lists
    1.99
     listing
    1.85
     checklist
    1.82
    Activations Density 0.025%

    No Known Activations