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. 17693
    Prev
    Next
    INDEX
    Explanations

    text marked with the notation "sic," indicating a perceived error or issue

    oai_token-act-pair · gpt-3.5-turbo

    references to musical notation or terms associated with music

    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
     divided
    -0.64
     mixed
    -0.61
     fet
    -0.60
     frequently
    -0.59
     fixed
    -0.57
     tightly
    -0.57
     dividing
    -0.57
     midway
    -0.56
     broke
    -0.56
     gradually
    -0.56
    POSITIVE LOGITS
    sic
    5.19
    emphasis
    1.33
    ?]
    1.25
    â̦]
    1.14
    !]
    1.13
    REDACTED
    1.13
    SPONSORED
    1.12
    â̦)
    1.01
    Laughs
    1.01
    laughs
    1.01
    Activations Density 0.017%

    No Known Activations