© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 14765
    Prev
    Next
    INDEX
    Explanations

    mentions of numbers at the beginning of phrases, potentially related to rankings or quantities

    oai_token-act-pair · gpt-3.5-turbo

    expressions of strong opinions or sentiments

    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
    economic
    -0.63
     Government
    -0.63
     Welfare
    -0.62
     welfare
    -0.62
     administr
    -0.61
     preventive
    -0.60
     withdrawing
    -0.60
     lawful
    -0.59
     unlawfully
    -0.59
     Employ
    -0.59
    POSITIVE LOGITS
     cinematic
    0.80
     sequels
    0.76
     Collider
    0.74
     hilar
    0.73
     cameo
    0.73
     teased
    0.73
     soundtrack
    0.72
     laughs
    0.71
     anthology
    0.70
     premie
    0.70
    Activations Density 3.160%

    No Known Activations