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. 1-RES-JB
    6. 19228
    Prev
    Next
    INDEX
    Explanations

    mentions of significant changes or shifts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "shift"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gom
    -0.76
    Interstitial
    -0.76
    amina
    -0.73
     Seym
    -0.71
     Printed
    -0.69
    DOC
    -0.67
    nam
    -0.67
    Blog
    -0.66
     External
    -0.65
    EVA
    -0.65
    POSITIVE LOGITS
     shift
    1.35
     shifts
    1.23
    shift
    1.06
     shifting
    1.03
     shif
    0.98
     Shift
    0.97
     shifted
    0.88
    iple
    0.86
     perspect
    0.82
    otom
    0.82
    Activations Density 0.012%

    No Known Activations