INDEX

Explanations

words related to emotions and sentiments, particularly with a focus on strong emotions

oai_token-act-pair · gpt-3.5-turbo

words related to emergency situations or crises

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SourceFile

-0.78

bed

-0.64

AVG

-0.64

 steroids

-0.64

 Toad

-0.61

 acids

-0.60

 Strauss

-0.58

 Rockies

-0.58

WIND

-0.58

 dreaded

-0.58

POSITIVE LOGITS

issions

1.17

peror

1.15

otional

1.15

achine

1.14

otions

1.10

igrant

1.07

igrants

1.06

ple

1.05

brace

1.05

phasis

1.03

Activations Density 0.025%

words related to emotions and sentiments, particularly with a focus on strong emotions

words related to emergency situations or crises

No Comments

No Known Activations