© 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. 6142
    Prev
    Next
    INDEX
    Explanations

    words related to official certifications or orders

    oai_token-act-pair · gpt-3.5-turbo

    terms related to certification and regulatory processes

    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
    finder
    -0.81
    birds
    -0.80
    breakers
    -0.70
    в
    -0.70
     Trib
    -0.67
    share
    -0.67
    haus
    -0.66
    bones
    -0.66
    ttes
    -0.66
    ãĤ¡
    -0.65
    POSITIVE LOGITS
    IFIED
    1.03
    INO
    0.95
    ECH
    0.93
    ITH
    0.92
    ARP
    0.91
    ulously
    0.90
    IFIC
    0.87
    ICAL
    0.87
    ION
    0.84
    wered
    0.83
    Activations Density 0.011%

    No Known Activations