© 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. 8-RES-JB
    11. 20507
    Prev
    Next
    INDEX
    Explanations

    short phrases or sentences that are abruptly cut off

    oai_token-act-pair · gpt-3.5-turbo

    punctuation markers, particularly periods and question marks

    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
    uers
    -0.71
     furnace
    -0.70
    ner
    -0.67
    itiz
    -0.66
     manageable
    -0.66
    din
    -0.66
    roud
    -0.66
    ratulations
    -0.65
    oing
    -0.65
     eligibility
    -0.65
    POSITIVE LOGITS
     Anyway
    1.47
     Secondly
    1.15
     Nevertheless
    1.10
     But
    1.06
     Suppose
    1.06
    Anyway
    1.06
     Nonetheless
    1.04
     Regardless
    1.04
     Especially
    1.01
     However
    0.99
    Activations Density 0.774%

    No Known Activations