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. 0-RES-JB
    6. 19644
    Prev
    Next
    INDEX
    Explanations

    instances of the word "votes"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "notes" and its variations in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     construction
    -0.60
    fine
    -0.59
     foundation
    -0.58
     mutual
    -0.58
     Inter
    -0.57
     joint
    -0.57
     Prince
    -0.57
     concurrent
    -0.57
     barrier
    -0.56
     DM
    -0.56
    POSITIVE LOGITS
    otes
    4.74
    oting
    2.41
    ote
    2.39
    oted
    2.09
    OTE
    2.06
    oters
    1.54
    otic
    1.49
    ots
    1.46
    otal
    1.41
    ota
    1.38
    Activations Density 0.010%

    No Known Activations