Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 10-RES-JB
    6. 19701
    Prev
    Next
    INDEX
    Explanations

    the word "z" at the end of certain entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter 'z'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Scots
    -0.71
     ransom
    -0.71
     Southeast
    -0.66
     entail
    -0.63
     behavi
    -0.63
     foss
    -0.60
     Malays
    -0.56
     butt
    -0.56
     Secret
    -0.56
     Cheong
    -0.56
    POSITIVE LOGITS
    ombie
    1.40
    ewski
    1.21
    enegger
    1.21
    ombies
    1.16
    alez
    1.10
    ealous
    1.07
    ymes
    1.06
    ebra
    1.05
    gerald
    1.03
    arella
    1.00
    Activations Density 0.064%

    No Known Activations