© 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. 707
    Prev
    Next
    INDEX
    Explanations

    terms related to specific cultural references or proper nouns

    oai_token-act-pair · gpt-3.5-turbo

    titles or names of fictional works and political references

    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
     FAR
    -0.56
     impunity
    -0.54
     reliably
    -0.52
     TEST
    -0.51
     captcha
    -0.50
     defense
    -0.49
     fertile
    -0.49
     overcrowd
    -0.49
     haste
    -0.49
     caps
    -0.48
    POSITIVE LOGITS
    foundland
    0.86
    odore
    0.78
    bye
    0.77
    itage
    0.77
    anmar
    0.75
    iversary
    0.73
    dinand
    0.72
    resa
    0.72
    ghazi
    0.71
    ablishment
    0.70
    Activations Density 0.536%

    No Known Activations