© 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. 12314
    Prev
    Next
    INDEX
    Explanations

    examples of actions or behaviors

    oai_token-act-pair · gpt-3.5-turbo

    references to individual or collective actions and their consequences

    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
    bid
    -0.71
     Moor
    -0.62
    population
    -0.62
     Dise
    -0.62
    ickets
    -0.61
    occup
    -0.60
    used
    -0.59
    orf
    -0.59
    MAL
    -0.58
    inately
    -0.58
    POSITIVE LOGITS
    uations
    1.08
     actions
    0.97
     ACTIONS
    0.96
    hops
    0.92
    uation
    0.85
    uits
    0.83
    bucks
    0.83
    uary
    0.80
    uggest
    0.80
    terday
    0.79
    Activations Density 0.021%

    No Known Activations