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. 8-RES-JB
    6. 3599
    Prev
    Next
    INDEX
    Explanations

    phrases related to familial relationships, especially those involving a brother-in-law

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to familial relationships and connections, particularly in the context of laws or legal matters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ramid
    -0.65
    Reviewer
    -0.63
    yip
    -0.63
     malf
    -0.60
    etheless
    -0.60
     Continent
    -0.59
    casters
    -0.59
    ourge
    -0.56
     operators
    -0.56
     funer
    -0.55
    POSITIVE LOGITS
     guise
    0.92
     vein
    0.81
     context
    0.80
     shape
    0.79
    shape
    0.78
     haste
    0.77
    depth
    0.76
     midst
    0.74
    Depth
    0.73
    clusive
    0.73
    Activations Density 0.192%

    No Known Activations