Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 18857
    Prev
    Next
    INDEX
    Explanations

    references to pages within documents

    oai_token-act-pair · gpt-3.5-turbo

    references to numbered pages in documents

    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
     Osiris
    -0.81
    oÄŁ
    -0.80
    Zone
    -0.76
    DERR
    -0.71
    rats
    -0.71
    opes
    -0.69
    roo
    -0.69
    âĹ¼
    -0.69
     Yel
    -0.68
    Allah
    -0.67
    POSITIVE LOGITS
     documents
    1.02
     document
    1.02
     manuscript
    0.99
     handwritten
    0.97
     detailing
    0.94
     papers
    0.91
     booklet
    0.91
     outlining
    0.90
     memos
    0.86
     dossier
    0.84
    Activations Density 0.038%

    No Known Activations