INDEX

Explanations

phrases prompting to read further for additional information

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Read" indicating calls to action or links to additional content

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

 towed

-0.71

Ã±o

-0.65

IDS

-0.64

opard

-0.63

urdue

-0.62

Whoever

-0.62

amel

-0.61

WC

-0.61

 ILCS

-0.60

ascal

-0.59

POSITIVE LOGITS

 aloud

1.09

dress

1.01

just

0.92

Write

0.92

iness

0.84

ers

0.82

ying

0.81

andowski

0.81

ership

0.79

me

0.76

Activations Density 0.034%

phrases prompting to read further for additional information

instances of the word "Read" indicating calls to action or links to additional content

No Comments

No Known Activations