INDEX

Explanations

mentions of staff members or reporters

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Staff."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

wit

-0.79

mobi

-0.79

BIP

-0.70

ICAN

-0.67

 Roosevelt

-0.66

FINE

-0.66

upon

-0.64

DonaldTrump

-0.64

Privacy

-0.64

selves

-0.63

POSITIVE LOGITS

 Staff

1.30

Staff

1.16

andowski

1.09

staff

1.05

 staff

1.00

eln

0.91

iciency

0.82

inguished

0.78

indal

0.78

 Writer

0.78

Activations Density 0.011%

mentions of staff members or reporters

occurrences of the word "Staff."

No Comments

No Known Activations