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. 11-RES-JB
    6. 3266
    Prev
    Next
    INDEX
    Explanations

    characters that are not recognized as regular text characters, such as "Ċ"

    oai_token-act-pair · gpt-3.5-turbo

    contexts or phrases indicating urgency or significant events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     eleph
    -0.88
     oun
    -0.74
     newsp
    -0.73
     tremend
    -0.71
     exha
    -0.70
     occas
    -0.68
     newcom
    -0.65
     undermin
    -0.61
     offending
    -0.60
     unprepared
    -0.60
    POSITIVE LOGITS
    ↵
    0.89
    GREEN
    0.87
    WASHINGTON
    0.86
    hello
    0.84
    ILLE
    0.82
    CLE
    0.82
    PROV
    0.80
    Updated
    0.79
    DEN
    0.77
    Authorities
    0.76
    Activations Density 0.192%

    No Known Activations