INDEX

Explanations

proper nouns, likely entities such as companies, people, and places

oai_token-act-pair · gpt-3.5-turbo

empty or unstructured text segments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thereof

-0.72

 thereto

-0.64

])

-0.63

Î¿

-0.62

Magikarp

-0.61

Â·Â·

-0.61

GBT

-0.60

GF

-0.60

 ..."

-0.60

POSITIVE LOGITS

theless

0.85

anyahu

0.80

withstanding

0.79

odore

0.76

resa

0.76

ashtra

0.75

sonian

0.73

xiety

0.72

ogether

0.69

asionally

0.69

Activations Density 0.220%

proper nouns, likely entities such as companies, people, and places

empty or unstructured text segments

No Comments

No Known Activations