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. 4228
    Prev
    Next
    INDEX
    Explanations

    words related to a negative or degrading description or action

    oai_token-act-pair · gpt-3.5-turbo

    words associated with the concept of meaning

    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
     conspiracy
    -0.68
     security
    -0.67
     Chicago
    -0.65
     prov
    -0.65
     USA
    -0.64
     Hawk
    -0.64
     Scotland
    -0.60
     Pitt
    -0.59
    stab
    -0.59
    Tex
    -0.59
    POSITIVE LOGITS
    aning
    4.47
    aned
    2.66
    anes
    1.38
    ans
    1.28
    asures
    1.18
    anism
    1.18
    an
    1.09
    anking
    1.07
    ANS
    1.05
    ane
    1.04
    Activations Density 0.012%

    No Known Activations