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. 3-RES-JB
    6. 15020
    Prev
    Next
    INDEX
    Explanations

    scientific or technical terms and abbreviations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sc" in various contexts, likely indicating a focus on themes of scrutiny or search

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Downloadha
    -0.92
    heit
    -0.85
    éĹĺ
    -0.76
    elsen
    -0.73
     stale
    -0.72
     remembrance
    -0.71
     lapse
    -0.70
    emonium
    -0.69
    hyde
    -0.68
     Archdemon
    -0.68
    POSITIVE LOGITS
    ammers
    1.20
    oops
    1.18
    opes
    1.18
    rupulous
    1.17
    ouring
    1.15
    rawling
    1.14
    aped
    1.13
    apers
    1.13
    urry
    1.11
    ape
    1.11
    Activations Density 0.008%

    No Known Activations