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. 3672
    Prev
    Next
    INDEX
    Explanations

    dates mentioned in the year 2019

    oai_token-act-pair · gpt-3.5-turbo

    references to the year 2019

    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
     pat
    -0.69
     marines
    -0.68
     contr
    -0.65
     leveled
    -0.64
     interrogation
    -0.59
    ax
    -0.59
     Lys
    -0.58
     mob
    -0.57
     resorted
    -0.57
    arm
    -0.56
    POSITIVE LOGITS
     2019
    3.82
     2021
    2.73
     2020
    2.61
     2018
    2.58
     2022
    2.50
    2019
    2.41
     2024
    2.29
     2025
    2.22
     2030
    2.02
    2018
    1.97
    Activations Density 0.007%

    No Known Activations