© 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. 8-RES-JB
    6. 7087
    Prev
    Next
    INDEX
    Explanations

    instances of the word "clear" followed by a number (e.g., "clear 10")

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating clarity or obviousness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tremend
    -0.95
    inse
    -0.77
    uld
    -0.77
     Loft
    -0.76
    ITAL
    -0.75
    FactoryReloaded
    -0.69
    eatures
    -0.69
     arrog
    -0.67
     nostalg
    -0.66
     destro
    -0.66
    POSITIVE LOGITS
    ances
    1.26
    cut
    1.16
    ance
    0.97
    cuts
    0.94
    headed
    0.93
    iary
    0.91
     indication
    0.89
    cutting
    0.85
     sailing
    0.84
     deline
    0.81
    Activations Density 0.045%

    No Known Activations