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. 0-RES-JB
    6. 3867
    Prev
    Next
    INDEX
    Explanations

    HTML tags and attributes

    oai_token-act-pair · gpt-3.5-turbo

    HTML tag attributes related to dimensions and sources

    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
     Garry
    -0.57
     timing
    -0.57
     Branch
    -0.56
     Kaz
    -0.55
     Harbour
    -0.55
     reception
    -0.55
     assisted
    -0.54
     hand
    -0.53
     Sutherland
    -0.52
     Kap
    -0.52
    POSITIVE LOGITS
    ="
    3.79
    ='
    2.83
    ="/
    2.79
    =""
    2.70
    ="#
    2.55
    =\"
    2.27
    ={
    2.12
    =$
    2.12
    =/
    1.97
    =(
    1.84
    Activations Density 0.010%

    No Known Activations