© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 23652
    Prev
    Next
    INDEX
    Explanations

    mentions of events or actions that are happening in anticipation of a particular future event

    oai_token-act-pair · gpt-3.5-turbo

    references to future events or actions

    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
    cu
    -0.73
     mart
    -0.73
    mini
    -0.73
    lee
    -0.72
    rug
    -0.71
    odor
    -0.68
    ASE
    -0.65
    izards
    -0.64
    bury
    -0.64
    zip
    -0.64
    POSITIVE LOGITS
     ahead
    0.88
    olicy
    0.78
    manship
    0.76
     antioxid
    0.74
    tains
    0.69
    itism
    0.69
    Ħ¢
    0.68
    ģĸ
    0.68
     Skip
    0.67
    itiz
    0.67
    Activations Density 0.011%

    No Known Activations