INDEX

Explanations

words related to geographical directions, especially 'east'

oai_token-act-pair · gpt-3.5-turbo

references to the "east" direction in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

INAL

-0.83

TABLE

-0.78

andom

-0.77

vous

-0.74

argon

-0.73

arantine

-0.72

abulary

-0.72

 Likes

-0.71

Number

-0.71

utable

-0.70

POSITIVE LOGITS

ward

1.33

side

1.10

wards

1.06

 coast

1.00

bound

0.98

west

0.88

front

0.87

 side

0.84

urst

0.83

east

0.81

Activations Density 0.016%

words related to geographical directions, especially 'east'

references to the "east" direction in various contexts

No Comments

No Known Activations