© 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. 15860
    Prev
    Next
    INDEX
    Explanations

    the letter "z" when it appears in the middle or at the end of words

    oai_token-act-pair · gpt-3.5-turbo

    references to the letter "z."

    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
    illet
    -0.69
     Merrill
    -0.68
     Stub
    -0.68
    theless
    -0.64
    isSpecialOrderable
    -0.64
    idate
    -0.63
    lishes
    -0.63
    ħĭ
    -0.63
    ivalry
    -0.62
    aints
    -0.62
    POSITIVE LOGITS
     z
    3.41
     Z
    1.95
    Z
    1.74
     ze
    1.70
     zo
    1.59
    z
    1.53
     Zo
    1.45
     zipper
    1.43
     Zer
    1.38
     zip
    1.37
    Activations Density 0.016%

    No Known Activations