INDEX

Explanations

large numbers or numerical values

oai_token-act-pair · gpt-3.5-turbo

references to image thumbnails and associated metadata

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pupp

-0.89

Wil

-0.87

Mus

-0.83

 Pascal

-0.81

Pear

-0.79

Pug

-0.78

-0.77

ait

-0.77

 brace

-0.75

Blake

-0.75

POSITIVE LOGITS

1.41

1.36

1.13

 Twelve

1.09

 twelve

1.09

1.06

XII

1.06

1.02

0.99

0.94

Activations Density 0.326%

large numbers or numerical values

references to image thumbnails and associated metadata

No Comments

No Known Activations