INDEX

Explanations

quotes or statements ending with a period

oai_token-act-pair · gpt-3.5-turbo

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

 totem

-0.82

 delinquent

-0.79

 nipple

-0.71

 undet

-0.70

 fermented

-0.70

 plent

-0.69

 transact

-0.69

 unsus

-0.68

 jugg

-0.68

 tampering

-0.68

POSITIVE LOGITS

 "...

1.20

 "âĢ¦

1.11

"(

0.97

 Asked

0.97

 However

0.96

 Adds

0.96

"'

0.95

 Meanwhile

0.94

<|endoftext|>

0.94

"[

0.92

Activations Density 0.184%

quotes or statements ending with a period

No Comments

No Known Activations