INDEX

Explanations

phrases indicating a person's quotes or reported statements

oai_token-act-pair · gpt-3.5-turbo

action verbs that indicate reporting or stating information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ffield

-0.63

iencies

-0.59

aughtered

-0.55

ashtra

-0.55

prus

-0.54

wcsstore

-0.52

estial

-0.51

 hires

-0.51

 occupancy

-0.51

 resemb

-0.51

POSITIVE LOGITS

 rhet

0.98

0.96

.[

0.91

!.

0.86

 sarcast

0.85

.(

0.84

 omin

0.83

.?

0.83

_.

0.81

().

0.81

Activations Density 0.103%

phrases indicating a person's quotes or reported statements

action verbs that indicate reporting or stating information

No Comments

No Known Activations