© 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. 1-RES-JB
    11. 4859
    Prev
    Next
    INDEX
    Explanations

    words related to the fictional character or concept called "Octopus"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific fictional or narrative titles and characters, particularly those associated with the word "opus."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uci
    -0.93
    umin
    -0.73
    uster
    -0.72
    credit
    -0.70
    icial
    -0.69
     electors
    -0.68
     elector
    -0.68
    hest
    -0.68
    ason
    -0.65
    luster
    -0.65
    POSITIVE LOGITS
    opus
    0.96
    ragon
    0.76
     Wonderland
    0.70
     Odyssey
    0.69
    aeda
    0.68
     Tosh
    0.68
    insula
    0.66
     Serpent
    0.66
    selage
    0.66
    asus
    0.65
    Activations Density 0.022%

    No Known Activations