© 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. 8-RES-JB
    11. 3275
    Prev
    Next
    INDEX
    Explanations

    past tense verbs related to physical actions or events

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "was" and its locations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inav
    -0.78
     entails
    -0.74
     Supports
    -0.72
     Highlights
    -0.69
     Previous
    -0.67
     Reduce
    -0.67
     Corruption
    -0.67
    osate
    -0.67
    icipated
    -0.66
     Critics
    -0.65
    POSITIVE LOGITS
     able
    1.21
     unable
    1.18
     unsure
    1.13
     amazed
    1.12
     afraid
    1.12
     adamant
    1.09
     fortunate
    1.06
     glad
    1.05
     lucky
    1.02
     greeted
    1.02
    Activations Density 0.347%

    No Known Activations