INDEX

Explanations

words related to causality or influence

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causation or effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ban

-0.65

thia

-0.65

 scrimmage

-0.62

 nurs

-0.60

---------

-0.59

-0.58

ASE

-0.57

br

-0.57

 pump

-0.56

 Witch

-0.56

POSITIVE LOGITS

hift

1.16

 sure

0.97

akable

0.81

paio

0.80

ensibly

0.77

ailable

0.75

enders

0.75

emort

0.74

ebin

0.74

rontal

0.73

Activations Density 0.120%

words related to causality or influence

phrases that indicate causation or effects

No Comments

No Known Activations