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. 20774
    Prev
    Next
    INDEX
    Explanations

    phrases or terms within square brackets

    oai_token-act-pair · gpt-3.5-turbo

    the closing brackets or brackets in the text

    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
     neighbors
    -0.52
     Twilight
    -0.51
     blight
    -0.50
     ro
    -0.49
     narrowly
    -0.48
     sa
    -0.47
     lo
    -0.45
     steel
    -0.45
     lur
    -0.45
     ever
    -0.45
    POSITIVE LOGITS
    ].
    3.76
    ]."
    3.17
     ].
    2.99
    ]).
    2.90
    ],
    2.77
    ];
    2.77
    ],"
    2.68
    .]
    2.65
    ]:
    2.51
    !]
    2.38
    Activations Density 0.008%

    No Known Activations