© 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. 3-RES-JB
    11. 15266
    Prev
    Next
    INDEX
    Explanations

    words that begin with the letter "E" followed by several letters, particularly names or the beginnings of longer words.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @danielepoterti

    words related to official titles and names

    oai_token-act-pair · gpt-3.5-turbo

    terms related to legal or formal 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
    anova
    -0.70
    itivity
    -0.64
    DAQ
    -0.61
    VW
    -0.59
    kov
    -0.58
     kittens
    -0.58
    isky
    -0.58
    natureconservancy
    -0.57
     MMR
    -0.56
    bg
    -0.56
    POSITIVE LOGITS
    tainment
    0.87
    hower
    0.81
    ionage
    0.77
    vironment
    0.77
    ascript
    0.77
    911
    0.75
    ilon
    0.74
    ghan
    0.70
    ploy
    0.65
    rahim
    0.65
    Activations Density 0.188%

    No Known Activations