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. 3-RES-JB
    6. 6863
    Prev
    Next
    INDEX
    Explanations

    references to a specific location or team, specifically related to Ipswich

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific location or entity related to "Ipswich."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    STEM
    -0.87
    WAYS
    -0.84
    LOAD
    -0.80
    Translation
    -0.73
    ARDS
    -0.71
    OTS
    -0.70
    Charge
    -0.70
    Redditor
    -0.69
    FUL
    -0.69
    士
    -0.67
    POSITIVE LOGITS
    wich
    1.16
    ilon
    1.03
     Ips
    0.95
    ilateral
    0.85
    olit
    0.85
    aret
    0.78
    ature
    0.77
    atell
    0.76
    ellig
    0.74
    borough
    0.73
    Activations Density 0.012%

    No Known Activations