INDEX

Explanations

phrases expressing caution or warning

oai_token-act-pair · gpt-3.5-turbo

expressions of caution or warning related to potential negative outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

utter

-0.70

found

-0.65

urch

-0.62

HH

-0.60

sb

-0.59

NH

-0.59

amb

-0.59

IENCE

-0.58

vation

-0.58

ains

-0.57

POSITIVE LOGITS

 lest

3.65

 Pastebin

1.07

 Canaver

0.82

 tremend

0.80

 holiest

0.79

soDeliveryDate

0.72

LET

0.71

 preferably

0.70

 dams

0.69

ĸļ

0.68

Activations Density 0.009%

phrases expressing caution or warning

expressions of caution or warning related to potential negative outcomes

No Comments

No Known Activations