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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 10570
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions such as folding, burden, and breathing

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to stability and structure

    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
     Pepe
    -0.89
     bot
    -0.73
     negro
    -0.71
    bay
    -0.70
    SPONSORED
    -0.66
    âĹ¼
    -0.65
    CVE
    -0.65
    Plex
    -0.65
     chats
    -0.64
    iques
    -0.64
    POSITIVE LOGITS
    olding
    2.89
    olded
    2.17
    inding
    1.90
    reath
    1.83
    urden
    1.77
    urry
    1.41
    enary
    1.34
    otive
    1.22
    undle
    1.19
    idden
    1.15
    Activations Density 0.040%

    No Known Activations