INDEX

Explanations

quotes starting with the word "that"

oai_token-act-pair · gpt-3.5-turbo

the word "that" used 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

aukee

-0.65

izont

-0.63

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.62

wn

-0.58

ãĥĺ

-0.57

WI

-0.57

ãĥ¥

-0.56

ãĥ¬

-0.55

BC

-0.55

 stall

-0.54

POSITIVE LOGITS

 although

0.82

esson

0.78

 fateful

0.77

soever

0.76

 they

0.72

 contradicts

0.72

chery

0.64

eday

0.64

ylum

0.63

"[

0.62

Activations Density 0.233%

quotes starting with the word "that"

the word "that" used in various contexts

No Comments

No Known Activations