Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 3-RES-JB
    6. 7312
    Prev
    Next
    INDEX
    Explanations

    words related to abstract concepts or emotions, such as empowerment, disappointment, fairness, doom, abundance, fulfillment, security, and so on

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express various forms of subjective perception or understanding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    theless
    -0.78
    sites
    -0.75
     downed
    -0.72
     helicop
    -0.71
    annis
    -0.69
     Cowboys
    -0.67
    onial
    -0.66
     prosecuted
    -0.63
     plaque
    -0.61
     Byrd
    -0.60
    POSITIVE LOGITS
    omatic
    0.90
    ively
    0.88
     Sense
    0.85
     sense
    0.82
    terday
    0.77
    uitive
    0.77
    ives
    0.76
    ual
    0.76
    ibly
    0.75
    ibility
    0.74
    Activations Density 0.021%

    No Known Activations