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

    Neuronpedia

    APICircuit 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. 16615
    Prev
    Next
    INDEX
    Explanations

    instances where something is being emphasized or pointed out

    oai_token-act-pair · gpt-3.5-turbo

    phrases that suggest causality or connection between concepts

    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
    ramids
    -0.68
    uku
    -0.66
    bill
    -0.65
    mill
    -0.65
    eer
    -0.61
    Cub
    -0.60
    jury
    -0.59
    Family
    -0.59
    ione
    -0.59
    case
    -0.58
    POSITIVE LOGITS
     mattered
    0.91
    rouse
    0.89
     distinguishes
    0.85
     bothers
    0.83
     prompted
    0.80
     determines
    0.79
     drew
    0.75
     attracts
    0.75
     inspires
    0.74
    ozy
    0.74
    Activations Density 0.119%

    No Known Activations