INDEX

Explanations

mentions of the word "vis" or "visit."

oai_token-act-pair · gpt-3.5-turbo

words related to visual representation or visibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Ĥª

-0.77

¶ħ

-0.76

RAW

-0.73

ropy

-0.67

anyahu

-0.66

CLR

-0.60

 Carnegie

-0.60

 overload

-0.60

 manners

-0.59

riad

-0.58

POSITIVE LOGITS

ibility

1.27

ibly

1.08

itors

1.08

ited

1.06

itation

0.87

itor

0.86

peed

0.86

idon

0.86

 versa

0.85

itism

0.85

Activations Density 0.013%

mentions of the word "vis" or "visit."

words related to visual representation or visibility

No Comments

No Known Activations