Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 14478
    Prev
    Next
    INDEX
    Explanations

    small parts or portions of larger quantities or entities

    oai_token-act-pair · gpt-3.5-turbo

    references to fractions or proportions 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
    esh
    -0.68
    intend
    -0.65
    went
    -0.64
    acho
    -0.64
     Helpful
    -0.63
     Pros
    -0.62
    love
    -0.62
     Regarding
    -0.60
    ESH
    -0.60
    atu
    -0.60
    POSITIVE LOGITS
     fraction
    3.75
     fractions
    2.69
     proportion
    1.44
     percentage
    1.32
     subset
    1.13
     portion
    1.10
     ratio
    1.10
     decimal
    1.04
     percentages
    1.03
     exponent
    1.01
    Activations Density 0.020%

    No Known Activations