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. 4252
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences related to technology and innovation

    oai_token-act-pair · gpt-3.5-turbo

    references to potential future developments and innovations

    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
     Maine
    -0.93
     phony
    -0.85
     knots
    -0.82
     shredded
    -0.80
     tossed
    -0.77
     scraps
    -0.77
     foul
    -0.76
     overturned
    -0.76
     improperly
    -0.76
     Arkansas
    -0.75
    POSITIVE LOGITS
    âĢ
    1.21
    ãĢIJ
    1.14
    ÅĤ
    1.08
    Laun
    1.08
    Ð
    1.06
    ãĢ
    1.03
    Äį
    1.03
     Ð
    1.02
    Development
    1.01
     Ä
    1.01
    Activations Density 0.706%

    No Known Activations