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. 4-RES-JB
    6. 21606
    Prev
    Next
    INDEX
    Explanations

    sequences of numbers in ascending order

    oai_token-act-pair · gpt-3.5-turbo

    numerical sequences or patterns associated with values

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

    No Comments

    Negative Logits
    afety
    -0.67
     shed
    -0.63
     Gle
    -0.61
    yre
    -0.59
    itch
    -0.57
    humane
    -0.57
    âĿ
    -0.56
     depl
    -0.55
    FUL
    -0.55
     condol
    -0.55
    POSITIVE LOGITS
    itate
    0.71
    odo
    0.67
    elo
    0.64
    aints
    0.64
    pei
    0.63
    ydia
    0.62
    ams
    0.61
    =-=-=-=-
    0.60
    aucus
    0.59
     Phar
    0.59
    Activations Density 0.079%

    No Known Activations