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. 7-RES-JB
    6. 16284
    Prev
    Next
    INDEX
    Explanations

    words related to causality and actions in past scenarios

    oai_token-act-pair · gpt-3.5-turbo

    references to the pronoun "it" and quantifiers in a variety of contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Making
    -0.63
    tern
    -0.62
     Hick
    -0.60
     Pants
    -0.59
    lash
    -0.59
     Comfort
    -0.58
    bell
    -0.57
     Invisible
    -0.57
     Marion
    -0.57
     Ark
    -0.56
    POSITIVE LOGITS
     evidenced
    0.90
     actionGroup
    0.89
     Âł Âł Âł Âł
    0.79
     [|
    0.78
     wont
    0.78
     Âł Âł Âł Âł Âł Âł Âł Âł
    0.75
     previously
    0.71
    çͰ
    0.70
     attest
    0.67
     ());
    0.66
    Activations Density 0.186%

    No Known Activations