© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 6524
    Prev
    Next
    INDEX
    Explanations

    titles of movies

    oai_token-act-pair · gpt-3.5-turbo

    opening parentheses used in various contexts, possibly related to titles or categorizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     impeachment
    -0.72
     electr
    -0.72
     handlers
    -0.70
     inund
    -0.67
     comparisons
    -0.64
     nitrogen
    -0.64
     unexpectedly
    -0.63
     irregularities
    -0.63
     ripe
    -0.63
     firmly
    -0.63
    POSITIVE LOGITS
    formerly
    1.29
    ?)
    1.24
    optional
    1.22
    aka
    1.21
    via
    1.21
    ...)
    1.20
    excluding
    1.17
    feat
    1.16
    includes
    1.15
    click
    1.14
    Activations Density 0.158%

    No Known Activations