© 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. 22230
    Prev
    Next
    INDEX
    Explanations

    Roman numerals, specifically "III"

    oai_token-act-pair · gpt-3.5-turbo

    terms and concepts related to substances and their chemical properties

    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
     ast
    -0.70
     dash
    -0.70
     Forward
    -0.69
    actionDate
    -0.68
     NAV
    -0.61
     incorrect
    -0.61
     Finder
    -0.61
     False
    -0.61
    ensions
    -0.60
     avoidance
    -0.60
    POSITIVE LOGITS
    ickers
    1.62
    III
    1.58
     practically
    1.50
    clad
    1.21
    iii
    1.18
    arella
    1.11
    covered
    1.08
    athed
    1.03
    yrus
    0.94
    kaya
    0.93
    Activations Density 0.051%

    No Known Activations