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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 0-RES-JB
    6. 5484
    Prev
    Next
    INDEX
    Explanations

    detailed or thorough analysis and explanations

    oai_token-act-pair · gpt-3.5-turbo

    references to analysis or examination of topics in depth

    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
     prov
    -0.72
     substitute
    -0.65
     substitutes
    -0.64
     Swan
    -0.63
     dod
    -0.63
     volunteers
    -0.62
     cheer
    -0.62
     Mr
    -0.62
     domestic
    -0.62
     male
    -0.61
    POSITIVE LOGITS
    depth
    3.97
    Depth
    2.47
     depth
    2.04
     Depth
    1.97
    DEP
    1.44
    length
    1.36
    detail
    1.18
     depths
    1.18
    deep
    1.13
    radius
    1.13
    Activations Density 0.015%

    No Known Activations