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. 4-RES-JB
    6. 5572
    Prev
    Next
    INDEX
    Explanations

    words related to bands, such as "band" used in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to "bands" or "band-related" concepts

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

    No Comments

    Negative Logits
     IMAGES
    -0.72
     Hus
    -0.62
    aminer
    -0.58
    MAL
    -0.56
     STA
    -0.56
    ensional
    -0.56
    matter
    -0.56
     HOU
    -0.55
    bably
    -0.55
     CONTROL
    -0.55
    POSITIVE LOGITS
    width
    1.08
    ictionary
    1.08
    band
    0.99
     bands
    0.98
    leader
    0.97
    bands
    0.96
    wagon
    0.95
    camp
    0.92
     band
    0.92
    ages
    0.92
    Activations Density 0.016%

    No Known Activations