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. 8-RES-JB
    6. 9720
    Prev
    Next
    INDEX
    Explanations

    numeric values in text

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating notable physical structures or objects

    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
    afety
    -0.63
    vironment
    -0.62
    abases
    -0.60
     contingency
    -0.59
     sqor
    -0.59
     Policies
    -0.58
    humans
    -0.57
    posts
    -0.57
    ©¶æ¥µ
    -0.57
    ships
    -0.57
    POSITIVE LOGITS
     Shutterstock
    0.66
     symbol
    0.64
     adorned
    0.60
     emblem
    0.59
     Doodle
    0.56
     prest
    0.56
     reminiscent
    0.55
     pierced
    0.55
     haunting
    0.54
     pier
    0.54
    Activations Density 1.496%

    No Known Activations