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. 8-RES-JB
    6. 5795
    Prev
    Next
    INDEX
    Explanations

    phrases contrasting different aspects or perspectives of a situation

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize the word "the" in comparison statements

    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
     Expend
    -0.64
    may
    -0.60
    awar
    -0.60
     SHARES
    -0.57
    Sense
    -0.55
    Isn
    -0.55
    ARE
    -0.54
     Annotations
    -0.54
     WATCHED
    -0.54
    aston
    -0.54
    POSITIVE LOGITS
     usual
    0.84
     mere
    0.80
     actual
    0.78
     necessarily
    0.76
     merely
    0.76
     ones
    0.74
    ividually
    0.70
     outright
    0.69
     traditional
    0.69
     slightest
    0.68
    Activations Density 0.138%

    No Known Activations