© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4-RES-JB
    11. 6121
    Prev
    Next
    INDEX
    Explanations

    words related to reading and written text

    oai_token-act-pair · gpt-3.5-turbo

    references to personal experiences and emotions

    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
     respectively
    -0.62
     themselves
    -0.61
     Beir
    -0.54
     apiece
    -0.49
    arettes
    -0.49
     liabilities
    -0.48
     Humph
    -0.48
     Belarus
    -0.48
     Hezbollah
    -0.48
     2020
    -0.47
    POSITIVE LOGITS
     myself
    1.19
     my
    0.79
    poke
    0.63
     blogging
    0.58
    VIDIA
    0.56
     crochet
    0.54
    eah
    0.53
    ograp
    0.53
    aido
    0.52
    oshop
    0.52
    Activations Density 1.609%

    No Known Activations