INDEX

Explanations

phrases expressing strong opinions or judgments

oai_token-act-pair · gpt-3.5-turbo

the contraction "you're" or its variations indicating the second person

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

rates

-0.67

obbies

-0.65

 implements

-0.64

ESE

-0.64

IDs

-0.64

 occurs

-0.62

ilts

-0.62

igraph

-0.61

kas

-0.61

UTERS

-0.61

POSITIVE LOGITS

 gonna

1.08

 welcome

1.00

 obligated

1.00

 liable

0.94

 probably

0.93

 likely

0.92

 lucky

0.92

 able

0.91

 entitled

0.90

 aware

0.89

Activations Density 0.180%

phrases expressing strong opinions or judgments

the contraction "you're" or its variations indicating the second person

No Comments

No Known Activations