© 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. 8-RES-JB
    11. 6169
    Prev
    Next
    INDEX
    Explanations

    phrases related to software programming and coding

    oai_token-act-pair · gpt-3.5-turbo

    references to gameplay experiences and actions

    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
    Not in Any Lists

    No Comments

    Negative Logits
    chell
    -0.67
    yss
    -0.63
    estern
    -0.63
    ppo
    -0.62
    orkshire
    -0.61
    nell
    -0.58
    lict
    -0.58
    tackle
    -0.58
    hess
    -0.57
    vel
    -0.55
    POSITIVE LOGITS
     thereafter
    1.14
     afterward
    0.87
     ensuing
    0.87
     afterwards
    0.82
     subsequent
    0.77
    )).
    0.70
     later
    0.70
     srfAttach
    0.69
     accordingly
    0.67
     thereto
    0.67
    Activations Density 1.010%

    No Known Activations