INDEX

Explanations

instances of the word "of" in the text

oai_token-act-pair · gpt-3.5-turbo

phrases referring to forms or types of something

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iets

-0.88

urers

-0.78

 WATCHED

-0.77

doms

-0.77

esses

-0.76

wives

-0.76

 Kingdoms

-0.73

heads

-0.72

kees

-0.71

estones

-0.69

POSITIVE LOGITS

 discrimination

0.85

 accommodation

0.85

 harassment

0.84

 oppression

0.83

 transportation

0.82

 insanity

0.81

 warfare

0.81

 coercion

0.76

 sabotage

0.75

 activism

0.75

Activations Density 0.067%

instances of the word "of" in the text

phrases referring to forms or types of something

No Comments

No Known Activations