INDEX

Explanations

phrases introducing new information or events

oai_token-act-pair · gpt-3.5-turbo

the word "that" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atre

-0.69

pling

-0.69

aukee

-0.63

ãĥ¥

-0.60

unic

-0.59

cosystem

-0.59

BC

-0.59

borg

-0.57

ãĥĺ

-0.57

Hok

-0.57

POSITIVE LOGITS

soever

0.92

 fateful

0.88

 although

0.74

 contradicts

0.73

 occurred

0.69

esson

0.69

eday

0.69

 arose

0.68

 Pastebin

0.67

"#

0.67

Activations Density 0.334%

phrases introducing new information or events

the word "that" in various contexts

No Comments

No Known Activations