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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 11251
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "Walwal", which could potentially be a name or entity

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Wal," likely indicating a focus on a specific proper noun or concept associated with that term

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nces
    -0.75
    ICLE
    -0.71
     Chaser
    -0.69
     Butterfly
    -0.64
     Webster
    -0.63
     Magicka
    -0.61
     Tune
    -0.60
     Buckley
    -0.59
     Addiction
    -0.59
     Vessel
    -0.59
    POSITIVE LOGITS
    wal
    1.06
    nut
    1.06
    mart
    1.02
    nuts
    1.00
    sten
    0.90
    ston
    0.89
    aby
    0.86
    ths
    0.86
    mere
    0.85
    nesday
    0.84
    Activations Density 0.008%

    No Known Activations