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. 7182
    Prev
    Next
    INDEX
    Explanations

    words related to subscribing or signing up for something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Sign" related to subscription calls to action

    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
    butt
    -0.68
    udeb
    -0.67
     wiped
    -0.66
     hunted
    -0.65
     craw
    -0.65
     recovering
    -0.62
    zhen
    -0.60
    ipel
    -0.60
     sucked
    -0.60
     boiled
    -0.59
    POSITIVE LOGITS
     Sign
    3.95
    Sign
    2.46
    sign
    2.41
     sign
    2.25
     SIGN
    2.08
     Signs
    1.96
    SIGN
    1.70
     signing
    1.63
     signs
    1.59
    signed
    1.44
    Activations Density 0.008%

    No Known Activations