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. 0-RES-JB
    6. 21937
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions of removal or elimination

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "wipe" and its variations in the context of elimination or erasure

    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
    Hamilton
    -0.77
     Hamilton
    -0.70
    vo
    -0.70
     Prospect
    -0.66
     IGF
    -0.65
     Kinnikuman
    -0.64
     trains
    -0.63
     Citation
    -0.62
     Barg
    -0.62
    chairs
    -0.62
    POSITIVE LOGITS
     wiped
    3.52
     wipe
    3.45
     wiping
    3.07
     wipes
    3.07
     erased
    1.67
     erase
    1.64
     eradicate
    1.42
     erad
    1.40
     scrub
    1.38
     extermin
    1.36
    Activations Density 0.024%

    No Known Activations