INDEX

Explanations

words that contain 'au'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "au"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ACTED

-0.76

EVA

-0.68

DonaldTrump

-0.67

 GOODMAN

-0.65

WHO

-0.64

STATE

-0.62

APS

-0.62

 Grimes

-0.61

frames

-0.61

âĸĪâĸĪâĸĪâĸĪ

-0.60

POSITIVE LOGITS

llah

1.10

gment

1.06

lette

0.93

clair

0.91

lly

0.88

cham

0.86

pload

0.85

ction

0.84

qua

0.84

fman

0.83

Activations Density 0.016%

words that contain 'au'

occurrences of the substring "au"

No Comments

No Known Activations