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. 6702
    Prev
    Next
    INDEX
    Explanations

    words related to the country "Transnistria"

    oai_token-act-pair · gpt-3.5-turbo

    references to geopolitical entities and trade agreements

    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
     doubtful
    -0.64
    bats
    -0.62
     subscribing
    -0.61
    axter
    -0.59
     spitting
    -0.59
    ouk
    -0.58
    arten
    -0.58
    sterdam
    -0.55
     reckoning
    -0.55
     weigh
    -0.55
    POSITIVE LOGITS
    Pacific
    0.92
    aucas
    0.84
    ropolitan
    0.83
    ional
    0.81
    Ń·
    0.75
     abdom
    0.71
    pter
    0.71
    ãĥķãĤ©
    0.70
    ildo
    0.70
    latable
    0.69
    Activations Density 0.084%

    No Known Activations