INDEX

Explanations

phrases that indicate comparison or contrast with the term "neither."

oai_token-act-pair · gpt-3.5-turbo

the word "neither" 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

Bang

-0.72

Ec

-0.71

ÙĴ

-0.71

uctions

-0.71

uers

-0.67

enos

-0.66

Vers

-0.66

Que

-0.66

soType

-0.64

ãĥķ

-0.63

POSITIVE LOGITS

theless

0.75

!--

0.75

 sexes

0.75

 overtly

0.72

llor

0.69

ndra

0.66

soever

0.66

 shalt

0.65

 sides

0.64

clud

0.63

Activations Density 0.009%

phrases that indicate comparison or contrast with the term "neither."

the word "neither" in various contexts

No Comments

No Known Activations