© 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. 11519
    Prev
    Next
    INDEX
    Explanations

    references to something being claimed, assumed, or rumored to be true

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "supposedly" to indicate speculation or uncertainty

    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
     Roof
    -0.75
    cest
    -0.74
    cair
    -0.70
    cloth
    -0.67
    oft
    -0.66
     hearts
    -0.65
    lag
    -0.65
    him
    -0.63
    lake
    -0.63
     Dragonbound
    -0.63
    POSITIVE LOGITS
    ãĥ¼ãĥĨãĤ£
    0.77
     contradict
    0.73
     mas
    0.69
     metic
    0.67
     belonging
    0.67
     proport
    0.66
     reside
    0.66
    ãĤ´ãĥ³
    0.66
     monopol
    0.66
     originate
    0.66
    Activations Density 0.013%

    No Known Activations