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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 3-RES-JB
    6. 1026
    Prev
    Next
    INDEX
    Explanations

    food-related terms, specifically relating to the size or cooking instructions

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "medium" related to various contexts or descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çīĪ
    -1.03
    ä½ľ
    -0.95
    èĥ
    -0.85
    zag
    -0.78
     Merch
    -0.77
    alog
    -0.76
    è¯
    -0.74
    çİĭ
    -0.74
    atche
    -0.74
    ãĤ¼
    -0.74
    POSITIVE LOGITS
     sized
    1.05
    medium
    0.87
    weights
    0.85
     humanoid
    0.81
    haired
    0.78
    ature
    0.77
    tackle
    0.75
    weight
    0.71
    heartedly
    0.71
     medium
    0.71
    Activations Density 0.008%

    No Known Activations