© 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. 11-RES-JB
    11. 2471
    Prev
    Next
    INDEX
    Explanations

    names of people or locations

    oai_token-act-pair · gpt-3.5-turbo

    single letters and specific sequences of letters, particularly those commonly found at the beginning of names or places

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     hoops
    -0.68
    kat
    -0.61
    aroo
    -0.58
    ilage
    -0.57
    taboola
    -0.55
     trig
    -0.55
    drawn
    -0.55
    achine
    -0.54
     bulletin
    -0.54
     caps
    -0.54
    POSITIVE LOGITS
    utsche
    1.01
    abase
    0.87
    shire
    0.78
    onso
    0.73
    stadt
    0.70
    byn
    0.70
    hyde
    0.69
    Ore
    0.69
    imil
    0.67
    ainment
    0.66
    Activations Density 0.209%

    No Known Activations