INDEX

Explanations

phrases related to supporting or endorsing a position

oai_token-act-pair · gpt-3.5-turbo

instances of the word "back" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

entric

-0.90

uyomi

-0.75

orp

-0.67

ities

-0.63

 ILCS

-0.61

æ©Ł

-0.61

å¥³

-0.61

risome

-0.60

uria

-0.59

 newsp

-0.57

POSITIVE LOGITS

track

1.20

stab

1.00

dated

0.97

packs

0.94

ped

0.91

dash

0.90

tracking

0.89

loaded

0.85

)=(

0.85

side

0.85

Activations Density 0.032%

phrases related to supporting or endorsing a position

instances of the word "back" and its various contexts

No Comments

No Known Activations