INDEX

Explanations

phrases that introduce or emphasize a following statement or observation

oai_token-act-pair · gpt-3.5-turbo

instances where the word "perhaps" is used

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ship

-0.83

ocaust

-0.76

ships

-0.76

iphate

-0.76

cium

-0.75

iak

-0.75

aturday

-0.74

bender

-0.72

cies

-0.72

akes

-0.72

POSITIVE LOGITS

 unsurprisingly

1.28

 sensing

0.97

 unsur

0.89

 exacerbated

0.87

 surprisingly

0.86

 reflecting

0.84

 unwittingly

0.84

 someday

0.83

 overlooked

0.83

 subconscious

0.81

Activations Density 0.051%

phrases that introduce or emphasize a following statement or observation

instances where the word "perhaps" is used

No Comments

No Known Activations