INDEX

Explanations

proper nouns related to specific locations, particularly "Hudson"

oai_token-act-pair · gpt-3.5-turbo

references to bodies of water, specifically Hudson and Thames rivers

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

ISTER

-0.74

ebook

-0.72

cious

-0.72

fully

-0.70

vol

-0.70

igned

-0.66

ion

-0.66

imated

-0.65

eper

-0.64

chio

-0.64

POSITIVE LOGITS

 Yards

1.00

 Valley

0.88

 Strait

0.86

 Hudson

0.77

Bay

0.75

Moh

0.73

é¾įåĸļå£«

0.71

 Falls

0.70

 Finch

0.69

 River

0.69

Activations Density 0.024%

proper nouns related to specific locations, particularly "Hudson"

references to bodies of water, specifically Hudson and Thames rivers

No Comments

No Known Activations