INDEX

Explanations

phrases indicating the quantity of two or pairs

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "two" and phrases indicating duality or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

earch

-0.69

enance

-0.61

alien

-0.58

ļ

-0.57

 snipp

-0.57

ILLE

-0.57

ordinate

-0.56

endo

-0.56

 lett

-0.55

eer

-0.55

POSITIVE LOGITS

 Both

1.30

one

1.26

Both

1.18

both

1.18

one

1.15

each

1.14

 Firstly

1.12

One

1.11

 Neither

1.11

 both

1.06

Activations Density 0.303%

phrases indicating the quantity of two or pairs

occurrences of the word "two" and phrases indicating duality or comparison

No Comments

No Known Activations