© 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. 3-RES-JB
    11. 1843
    Prev
    Next
    INDEX
    Explanations

    words related to political or social issues, often associated with criticism or opposition

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "ants" and its variations in texts

    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
    STEM
    -0.71
    FY
    -0.70
     Raleigh
    -0.68
    ãĤ¨ãĥ«
    -0.66
    SET
    -0.65
    Tx
    -0.65
    LOC
    -0.65
    ×ŀ
    -0.65
    pton
    -0.64
    \\\\\\\\
    -0.64
    POSITIVE LOGITS
    wana
    1.00
    ucker
    1.00
    ettings
    0.96
    terday
    0.91
    hips
    0.91
    heet
    0.89
    cha
    0.88
    lip
    0.87
    hell
    0.85
    asma
    0.84
    Activations Density 0.013%

    No Known Activations