INDEX

Explanations

the word "like" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "looks like."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

chin

-0.79

hiba

-0.75

ourse

-0.72

inion

-0.71

ulty

-0.70

angel

-0.70

iets

-0.70

inoa

-0.70

overy

-0.69

rax

-0.69

POSITIVE LOGITS

lihood

1.53

lier

1.05

liest

1.01

liness

0.87

 ours

0.83

terday

0.81

0.79

 minded

0.78

 hers

0.74

 parity

0.74

Activations Density 0.092%

the word "like" in various contexts

instances of the phrase "looks like."

No Comments

No Known Activations