INDEX

Explanations

words related to causation and outcome

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate consequences or effects

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

ĸļ

-0.73

Jur

-0.66

 Bastard

-0.62

Kun

-0.60

 Wend

-0.60

 Pants

-0.58

 Kush

-0.58

raint

-0.57

Sut

-0.57

 Sabb

-0.57

POSITIVE LOGITS

depending

0.80

 depending

0.78

 tricky

0.72

gettable

0.72

 safely

0.69

odder

0.68

ESE

0.68

 anywhere

0.67

 easily

0.64

pole

0.64

Activations Density 0.300%

words related to causation and outcome

phrases that indicate consequences or effects

No Comments

No Known Activations