© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 23884
    Prev
    Next
    INDEX
    Explanations

    names, likely of a specific person or a character

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ser" in various forms, likely searching for service-related or contextual references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Rated
    -0.75
     VIDE
    -0.66
     Ou
    -0.63
    Reviewer
    -0.62
    externalActionCode
    -0.62
     Downing
    -0.60
     Lovecraft
    -0.59
    endor
    -0.59
    AUT
    -0.58
    REF
    -0.58
    POSITIVE LOGITS
    geant
    1.26
    otonin
    0.98
    ser
    0.97
    vers
    0.96
    fen
    0.85
    vol
    0.85
    plings
    0.85
    pent
    0.84
    seless
    0.83
    ve
    0.81
    Activations Density 0.009%

    No Known Activations