INDEX

Explanations

phrases related to specific numerical sequences (e.g., "3rd and short")

oai_token-act-pair · gpt-3.5-turbo

instances of the word "and"

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

ãĤ©

-0.79

ONSORED

-0.70

achine

-0.70

exempt

-0.68

bo

-0.66

stals

-0.66

GD

-0.65

edIn

-0.64

ubi

-0.62

casts

-0.62

POSITIVE LOGITS

 foremost

0.99

 mids

0.75

rogen

0.73

 parcel

0.68

 limb

0.67

 consequently

0.67

rogens

0.67

 balcony

0.66

 shoulders

0.64

 vice

0.64

Activations Density 0.212%

phrases related to specific numerical sequences (e.g., "3rd and short")

instances of the word "and"

No Comments

No Known Activations