INDEX

Explanations

phrases expressing various degrees of intensity or emphasis, from mild to strong

oai_token-act-pair · gpt-3.5-turbo

instances of the character "L" or related phonetic sounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 mathemat

-0.79

 carbohyd

-0.75

geries

-0.73

 misunder

-0.72

ĵĺ

-0.71

 exha

-0.68

 Gutenberg

-0.68

 disadvant

-0.66

 pharmacy

-0.64

 Palestin

-0.64

POSITIVE LOGITS

ï¸ı

1.11

å¥³

0.81

ï¸

0.79

own

0.76

Â§

0.76

âĻ

0.73

Balt

0.73

Exit

0.73

Â£

0.72

0000000000000000

0.72

Activations Density 0.415%

phrases expressing various degrees of intensity or emphasis, from mild to strong

instances of the character "L" or related phonetic sounds

No Comments

No Known Activations