© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 2899
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Mem" or forms of it, such as "Memoir."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "memory" and variations of the word

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Baird
    -0.70
    Ö¼
    -0.69
     TT
    -0.69
    WB
    -0.69
     WB
    -0.68
    finder
    -0.67
    heit
    -0.65
     squared
    -0.65
     Browne
    -0.64
     enforcement
    -0.64
    POSITIVE LOGITS
     Mem
    3.97
    Mem
    3.11
    mem
    2.85
     mem
    2.24
     MEM
    1.99
     memes
    1.53
     Memor
    1.39
     Memory
    1.32
     meme
    1.32
     Memories
    1.22
    Activations Density 0.022%

    No Known Activations