© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4-RES-JB
    11. 16121
    Prev
    Next
    INDEX
    Explanations

    words related to hierarchical structures or subdivisions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "branch" and its variations in a variety of contexts

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

    No Comments

    Negative Logits
     Ukrain
    -0.71
    Org
    -0.67
    Iv
    -0.64
    ocker
    -0.64
    aving
    -0.63
    BAT
    -0.63
    ted
    -0.62
    ËĪ
    -0.62
    Ear
    -0.60
    forced
    -0.60
    POSITIVE LOGITS
     branches
    0.89
    illary
    0.85
    ranch
    0.85
     branch
    0.84
    dale
    0.83
    zhou
    0.81
    itect
    0.81
     Branch
    0.78
    tip
    0.78
    esian
    0.77
    Activations Density 0.011%

    No Known Activations