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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 4182
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences emphasizing extreme circumstances or situations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "extremely" and its context, often emphasizing intensity or rarity

    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
     Colleges
    -0.76
    lins
    -0.71
    sburgh
    -0.71
    ja
    -0.70
    agra
    -0.66
    lest
    -0.66
    iture
    -0.66
    ajo
    -0.65
    ta
    -0.65
     Cities
    -0.65
    POSITIVE LOGITS
     rare
    0.93
     expensive
    0.86
     inefficient
    0.84
     unlikely
    0.82
     difficult
    0.81
     unstable
    0.79
     sensitive
    0.78
     costly
    0.78
     unsu
    0.77
     beneficial
    0.77
    Activations Density 0.016%

    No Known Activations