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. 2943
    Prev
    Next
    INDEX
    Explanations

    phrases that involve recognizing or acknowledging something

    oai_token-act-pair · gpt-3.5-turbo

    words related to acknowledgment and awareness

    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
    colored
    -1.26
    defense
    -1.21
    avored
    -1.10
    --
    -1.08
     center
    -1.06
     --
    -1.04
     labor
    -1.04
     favor
    -1.03
     harbor
    -1.02
     honoring
    -1.02
    POSITIVE LOGITS
     realise
    2.70
     realised
    2.56
     recognise
    2.51
     organise
    2.40
     recognised
    2.27
     apologise
    2.23
     Whilst
    2.22
     organising
    2.18
     apologised
    2.15
     analyse
    2.14
    Activations Density 0.097%

    No Known Activations