© 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. 1-RES-JB
    11. 3467
    Prev
    Next
    INDEX
    Explanations

    instances of phrases denoting collaboration or coordination between entities

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the concept of connection or combination

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mbuds
    -0.73
    lying
    -0.66
    cas
    -0.64
     Sabha
    -0.62
     Citiz
    -0.62
    chat
    -0.62
     srfAttach
    -0.60
    cigarettes
    -0.60
    ambling
    -0.59
    ĸ
    -0.59
    POSITIVE LOGITS
    creen
    0.87
    ually
    0.85
    insula
    0.83
    ioned
    0.83
    ality
    0.83
     conjunction
    0.81
    swick
    0.78
    ãĥ¼ãĥĨ
    0.75
    ally
    0.74
    oise
    0.72
    Activations Density 0.007%

    No Known Activations