INDEX

Explanations

the word "second" followed by a noun or adjective

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of a "second" or additional opportunity or iteration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

IPS

-0.71

endas

-0.70

acters

-0.68

Marginal

-0.66

endo

-0.63

ARDS

-0.62

tics

-0.62

à©

-0.62

Reward

-0.62

Attributes

-0.62

POSITIVE LOGITS

 baseman

1.09

 consecutive

0.95

hand

0.89

 successive

0.87

worldly

0.85

 glance

0.84

 stint

0.83

arily

0.76

 round

0.75

 wave

0.75

Activations Density 0.055%

the word "second" followed by a noun or adjective

phrases related to the concept of a "second" or additional opportunity or iteration

No Comments

No Known Activations