© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 20012
    Prev
    Next
    INDEX
    Explanations

    the number "Nine" or variations of the word "nine"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "nine" in various contexts

    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
    rill
    -0.71
    BIP
    -0.68
    Agent
    -0.67
    Boost
    -0.63
    lance
    -0.63
     Uriel
    -0.63
    aminer
    -0.62
    spread
    -0.61
    immune
    -0.61
    feature
    -0.61
    POSITIVE LOGITS
    teen
    1.18
     months
    1.08
    teenth
    1.07
    een
    1.06
     hundred
    1.04
    eenth
    1.01
     weeks
    0.90
    eteen
    0.89
     years
    0.87
     consecutive
    0.85
    Activations Density 0.010%

    No Known Activations