© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 451
    Prev
    Next
    INDEX
    Explanations

    verbs or nouns related to the concept of importance or focus

    oai_token-act-pair · gpt-3.5-turbo

    key nouns and concepts related to significant actions or outcomes

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

    No Comments

    Negative Logits
    é»Ĵ
    -0.83
    respect
    -0.71
    DoS
    -0.70
    merce
    -0.69
    CRIP
    -0.68
    earances
    -0.67
    åĪ
    -0.65
    ãĥīãĥ©
    -0.65
    ENSE
    -0.64
    each
    -0.64
    POSITIVE LOGITS
    liest
    1.29
    iest
    1.20
     equivalent
    1.03
     centerpiece
    0.87
     antidote
    0.81
     safest
    0.79
     opener
    0.79
     happiest
    0.75
    hest
    0.75
     easiest
    0.75
    Activations Density 0.344%

    No Known Activations