INDEX

Explanations

phrases related to investigations and evaluations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" and related phrases suggestive of examination or investigation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

cil

-0.88

nb

-0.79

hover

-0.69

Alias

-0.64

Iterator

-0.63

leon

-0.63

uti

-0.63

Slot

-0.63

âĿ

-0.62

Enjoy

-0.61

POSITIVE LOGITS

 feasibility

1.25

 entire

1.20

 entirety

1.19

 contents

1.13

 possibility

1.10

 merits

1.08

 extent

1.06

 effectiveness

1.04

 situation

1.03

 legality

1.02

Activations Density 0.249%

phrases related to investigations and evaluations

instances of the word "the" and related phrases suggestive of examination or investigation

No Comments

No Known Activations