© 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. 0-RES-JB
    11. 12217
    Prev
    Next
    INDEX
    Explanations

    phrases or words related to appropriate actions or responses

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "appropriately" and related concepts that denote suitability or correctness

    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
    ertility
    -0.78
    poll
    -0.71
    hammad
    -0.70
     favour
    -0.69
    pict
    -0.67
    NetMessage
    -0.65
     contradictory
    -0.65
    rouch
    -0.61
     miracles
    -0.61
     intercourse
    -0.61
    POSITIVE LOGITS
     appropriately
    2.23
     aptly
    1.48
     underscore
    1.35
    appropriately
    1.30
     underscores
    1.24
     sensibilities
    1.09
    rats
    1.01
     responsibly
    0.94
     ens
    0.94
     notwithstanding
    0.92
    Activations Density 0.067%

    No Known Activations