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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 3507
    Prev
    Next
    INDEX
    Explanations

    concrete-related terms and materials

    oai_token-act-pair · gpt-3.5-turbo

    references to building materials and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Specific
    -0.88
     Kard
    -0.81
     Prevention
    -0.76
    Kids
    -0.75
    irlf
    -0.74
     Relations
    -0.72
    ivia
    -0.70
    yssey
    -0.70
     Predict
    -0.70
    YN
    -0.70
    POSITIVE LOGITS
     slab
    0.98
     rods
    0.97
    glass
    0.96
     alloy
    0.94
     sculpture
    0.92
     plaster
    0.92
     coffin
    0.91
     sculptures
    0.91
     clay
    0.91
     oak
    0.90
    Activations Density 0.138%

    No Known Activations