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. 16767
    Prev
    Next
    INDEX
    Explanations

    phrases related to predictions or expectations

    oai_token-act-pair · gpt-3.5-turbo

    sentences discussing potential outcomes or evaluations

    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
    illegal
    -0.68
    ãĤ½
    -0.66
    Downloadha
    -0.66
    âĢ
    -0.65
     Facts
    -0.62
    ãĤ®
    -0.62
     è£ıç
    -0.61
    Dear
    -0.61
     unlawfully
    -0.60
     unlawful
    -0.60
    POSITIVE LOGITS
     kinda
    0.97
     definitely
    0.94
     hopefully
    0.92
     mell
    0.91
     adapt
    0.90
     fare
    0.87
     still
    0.86
     synerg
    0.85
     complement
    0.84
     adjust
    0.84
    Activations Density 0.489%

    No Known Activations