© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 22996
    Prev
    Next
    INDEX
    Explanations

    words related to wheels or spinning

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Whe" and its variations used in different contexts

    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
    pmwiki
    -0.82
     Defenders
    -0.78
    EMBER
    -0.71
    ographers
    -0.70
     Echoes
    -0.67
    76561
    -0.66
     Gallery
    -0.66
    UAL
    -0.65
     Mour
    -0.64
    ocene
    -0.64
    POSITIVE LOGITS
     Whe
    1.07
    ldon
    0.93
    esome
    0.87
    esley
    0.86
    whe
    0.84
    eling
    0.84
    atson
    0.83
    iago
    0.82
    ys
    0.82
    yton
    0.81
    Activations Density 0.005%

    No Known Activations