© 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. 3242
    Prev
    Next
    INDEX
    Explanations

    instances of company names or proper nouns

    oai_token-act-pair · gpt-3.5-turbo

    commonly used phrases and conjunctions that indicate lists or examples

    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
    onym
    -0.63
    bucks
    -0.62
    irie
    -0.62
    hei
    -0.61
    ide
    -0.60
    sein
    -0.60
    sburg
    -0.60
    ths
    -0.59
    ei
    -0.58
    isse
    -0.58
    POSITIVE LOGITS
     which
    1.37
     whose
    1.32
    whose
    1.23
    which
    1.19
     where
    1.03
     wherein
    1.02
     whereby
    0.98
     who
    0.95
     although
    0.91
     aka
    0.91
    Activations Density 0.496%

    No Known Activations