© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 14219
    Prev
    Next
    INDEX
    Explanations

    verbs related to slow or difficult movement

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "crawl" and its variations

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

    No Comments

    Negative Logits
    Ľ
    -0.84
    zona
    -0.82
    lda
    -0.80
    xtap
    -0.77
    ces
    -0.74
    orical
    -0.73
    ī
    -0.72
    ����
    -0.72
    rament
    -0.72
    ulet
    -0.72
    POSITIVE LOGITS
     craw
    1.22
     crawl
    1.15
     Craw
    1.10
     crawling
    1.08
    halla
    0.90
     crawled
    0.90
    odge
    0.75
    rawling
    0.73
    vier
    0.72
    ingly
    0.71
    Activations Density 0.012%

    No Known Activations