© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 10373
    Prev
    Next
    INDEX
    Explanations

    the string "mi" with an activation value of 9 or 10

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "mi" in various contexts

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

    No Comments

    Negative Logits
    lain
    -0.92
    swick
    -0.89
    egg
    -0.83
    rooms
    -0.82
    tions
    -0.81
    IBLE
    -0.77
    glers
    -0.76
    fully
    -0.73
    lift
    -0.72
    space
    -0.71
    POSITIVE LOGITS
    ya
    0.91
    wi
    0.87
    oti
    0.87
    ño
    0.80
    otti
    0.80
    ovember
    0.80
    wana
    0.78
    omi
    0.75
    ña
    0.74
    opsis
    0.74
    Activations Density 0.010%

    No Known Activations