© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4020
    Prev
    Next
    INDEX
    Explanations

    words related to intensity or extremeness

    oai_token-act-pair · gpt-3.5-turbo

    words that denote a state or condition, particularly those ending with the suffix '-ous'

    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
     banned
    -0.61
     secretaries
    -0.60
     currencies
    -0.59
     toler
    -0.58
     Euros
    -0.58
     mirror
    -0.58
     lucid
    -0.57
     jeans
    -0.56
     stub
    -0.56
     yards
    -0.56
    POSITIVE LOGITS
    ous
    4.60
    ously
    3.30
    OUS
    2.66
    osity
    1.87
    ouse
    1.82
    ouses
    1.70
    oused
    1.66
    ious
    1.50
    uous
    1.42
    vous
    1.34
    Activations Density 0.011%

    No Known Activations