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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & 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. 8-RES-JB
    6. 6643
    Prev
    Next
    INDEX
    Explanations

    phrases related to decision-making or preferences

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing desires and intentions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •contextual features
    fires on two token verb pairs like "want to" "take it" "have some"

    No Comments

    Negative Logits
     Seym
    -0.74
     Argon
    -0.69
     Niet
    -0.64
     Cele
    -0.60
    esides
    -0.59
     vulner
    -0.59
    ãĥ¼ãĥĨ
    -0.59
     Vaugh
    -0.57
    edIn
    -0.57
     fortun
    -0.57
    POSITIVE LOGITS
    malink
    0.89
     largeDownload
    0.80
     ]
    0.70
     ];
    0.69
     ][
    0.69
    reply
    0.68
     Expand
    0.68
    acan
    0.67
     RELEASE
    0.64
    embed
    0.63
    Activations Density 0.375%

    No Known Activations