© 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. 0-RES-JB
    11. 2849
    Prev
    Next
    INDEX
    Explanations

    proper nouns or specific terms/phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate examples or specifications of previously mentioned concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tears
    -0.58
    ou
    -0.58
    ater
    -0.57
     next
    -0.56
    ickets
    -0.55
     luck
    -0.55
    im
    -0.55
    imeter
    -0.54
    ort
    -0.54
    imming
    -0.53
    POSITIVE LOGITS
     namely
    3.95
     viz
    2.53
     Firstly
    1.45
     secondly
    1.42
     notably
    1.42
     ie
    1.35
     Specifically
    1.28
     whereby
    1.26
     Including
    1.23
     albeit
    1.19
    Activations Density 0.009%

    No Known Activations