© 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. 0-RES-JB
    11. 20522
    Prev
    Next
    INDEX
    Explanations

    phrases containing specific words or names found in URLs or text

    oai_token-act-pair · gpt-3.5-turbo

    specific names or identifiers related to groups or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     brewed
    -0.72
     Templar
    -0.69
     Belgium
    -0.66
    Meat
    -0.64
     Template
    -0.64
     Fortune
    -0.63
     Tac
    -0.63
     Sandwich
    -0.62
     Subway
    -0.61
     Bat
    -0.61
    POSITIVE LOGITS
    hod
    4.55
    endez
    1.67
    hea
    1.38
    ha
    1.04
    ixon
    0.97
    aleigh
    0.96
    utic
    0.95
    ilee
    0.92
    utics
    0.92
    ovych
    0.89
    Activations Density 0.020%

    No Known Activations