© 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. 0-RES-JB
    6. 18028
    Prev
    Next
    INDEX
    Explanations

    keywords related to publications, specifically monthly and popular content

    oai_token-act-pair · gpt-3.5-turbo

    terms related to frequency or regular updates in content

    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
     twe
    -0.73
    steen
    -0.69
    ãĥ£
    -0.69
     lip
    -0.67
     blockers
    -0.67
    ĸļ
    -0.67
     ticking
    -0.66
    Transform
    -0.66
    ãĤ£
    -0.65
     goddamn
    -0.65
    POSITIVE LOGITS
     Monthly
    3.06
     Popular
    1.76
     Minor
    1.53
     Quarterly
    1.40
     Historical
    1.30
     Means
    1.26
    graduate
    1.18
     Mutual
    1.14
     Encyclopedia
    1.00
     Institution
    0.99
    Activations Density 0.049%

    No Known Activations