© 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. 4-RES-JB
    11. 13496
    Prev
    Next
    INDEX
    Explanations

    phrases related to continuation, quick actions, and important steps or measures

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating summaries or descriptions of various topics

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

    No Comments

    Negative Logits
    seys
    -0.89
    alties
    -0.84
    ween
    -0.83
    ngth
    -0.79
    oes
    -0.78
    dies
    -0.78
    aneers
    -0.75
    obiles
    -0.72
    bats
    -0.70
    lime
    -0.69
    POSITIVE LOGITS
     conjecture
    0.71
     blatant
    0.70
     bold
    0.68
     continuation
    0.68
     unintentional
    0.68
     prol
    0.68
     ado
    0.66
     moot
    0.66
     DAR
    0.65
     PvP
    0.65
    Activations Density 0.269%

    No Known Activations