© 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. 2386
    Prev
    Next
    INDEX
    Explanations

    comparative or superlative adjectives

    oai_token-act-pair · gpt-3.5-turbo

    the word "more" in various contexts

    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
     dispatch
    -0.62
    essed
    -0.61
     deduction
    -0.58
     camera
    -0.57
     session
    -0.57
     beast
    -0.56
     privilege
    -0.56
    ortion
    -0.55
     initial
    -0.55
     volt
    -0.53
    POSITIVE LOGITS
    more
    4.21
    MORE
    1.79
    better
    1.55
    More
    1.31
    mere
    1.29
    mont
    1.17
     More
    1.13
    much
    1.10
    many
    1.07
    higher
    1.06
    Activations Density 0.007%

    No Known Activations