INDEX

Explanations

the word "like" being used in a negative context

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "don't like" 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

alm

-0.80

vantage

-0.77

utical

-0.77

ositories

-0.76

TAIN

-0.75

rontal

-0.74

hern

-0.73

elin

-0.73

chin

-0.72

abetic

-0.72

POSITIVE LOGITS

lihood

1.07

nor

0.88

 anything

0.83

liness

0.83

 anybody

0.82

 anymore

0.79

ably

0.77

liest

0.75

 surprises

0.73

any

0.72

Activations Density 0.034%

the word "like" being used in a negative context

instances of the phrase "don't like" in various contexts

No Comments

No Known Activations