INDEX

Explanations

phrases related to the usage of various items or components in different contexts

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "in" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

llah

-0.75

ovych

-0.69

iable

-0.69

daq

-0.68

osal

-0.67

-0.66

rio

-0.65

ydia

-0.64

HAEL

-0.63

worthiness

-0.61

POSITIVE LOGITS

 conjunction

1.42

 lieu

1.21

 tandem

0.96

 emergencies

0.90

 constructing

0.83

 combination

0.82

 practice

0.82

ordinate

0.82

 moderation

0.80

 situ

0.80

Activations Density 0.187%

phrases related to the usage of various items or components in different contexts

the occurrence of the word "in" in various contexts

No Comments

No Known Activations