© 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 EvalsExportsinterp-engineNEWAPI 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. 4604
    Prev
    Next
    INDEX
    Explanations

    text encoded with special symbols such as 'Ċ' and 'ľ'

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating quantitative measures or statistical information

    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
     freel
    -0.59
     deceived
    -0.59
     undermin
    -0.57
    ulkan
    -0.57
     appe
    -0.56
     Saiyan
    -0.56
     assignment
    -0.55
    hement
    -0.55
     orchestr
    -0.55
    uca
    -0.55
    POSITIVE LOGITS
    ³³³³
    1.05
    ³³³³³³³³³³³³³³³³
    1.04
    ³³³³³³³³
    0.98
    Advertisement
    0.94
    Posted
    0.87
    TABLE
    0.84
    ³³³
    0.81
    posted
    0.81
    Newsletter
    0.80
    ³³
    0.80
    Activations Density 1.080%

    No Known Activations