© 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. 0-RES-JB
    6. 11800
    Prev
    Next
    INDEX
    Explanations

    phrases related to deceptive or unethical behaviors

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "dealing."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cluster
    -0.59
     astroph
    -0.58
    NEWS
    -0.58
     supers
    -0.58
     national
    -0.57
     chalk
    -0.57
     science
    -0.57
     precincts
    -0.55
     Delta
    -0.55
     Berry
    -0.55
    POSITIVE LOGITS
    ealing
    3.92
    alking
    1.30
    eeds
    1.28
    icably
    1.22
    eals
    1.07
    iless
    1.05
    orting
    0.99
    lying
    0.98
    olver
    0.96
    rieving
    0.96
    Activations Density 0.024%

    No Known Activations