© 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
    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. 20011
    Prev
    Next
    INDEX
    Explanations

    comparisons where one quantity is larger than another

    oai_token-act-pair · gpt-3.5-turbo

    references to quantities that surpass certain thresholds

    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
    tta
    -0.65
    iku
    -0.62
    ama
    -0.61
     Brief
    -0.61
     Templ
    -0.60
     Cooperative
    -0.60
    bas
    -0.60
    arty
    -0.59
     Talking
    -0.57
     Velvet
    -0.57
    POSITIVE LOGITS
     exceed
    3.46
     exceeded
    3.09
     exceeds
    2.82
     exceeding
    2.71
     surpass
    2.01
     surpassed
    1.98
     outweigh
    1.48
    ceed
    1.31
     outwe
    1.25
     excess
    1.24
    Activations Density 0.019%

    No Known Activations