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. 6-RES-JB
    6. 2109
    Prev
    Next
    INDEX
    Explanations

    phrases indicating actions or decisions to be taken

    oai_token-act-pair · gpt-3.5-turbo

    various forms of verbs and their implications in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Malays
    -0.62
     Malaysia
    -0.57
    Drag
    -0.52
    atl
    -0.51
    ©¶æ
    -0.50
    ships
    -0.49
    Shar
    -0.49
     Bild
    -0.48
     Lars
    -0.48
     KL
    -0.48
    POSITIVE LOGITS
     temptation
    0.67
     indistinguishable
    0.59
     endif
    0.55
    âĵĺ
    0.54
     elig
    0.54
    morrow
    0.54
    '."
    0.53
    .'"
    0.53
     goto
    0.52
    etheless
    0.52
    Activations Density 1.284%

    No Known Activations