INDEX

Explanations

phrases or words related to numerical values and statistics

oai_token-act-pair · gpt-3.5-turbo

key phrases and indicators of trends or changes over time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 âĢ

-0.86

-0.79

to

-0.78

âĢ

-0.77

and

-0.74

.—

-0.69

!!!!

-0.67

 âĨĴ

-0.67

................................

-0.67

——

-0.66

POSITIVE LOGITS

 pione

0.84

ü

0.72

rawdownload

0.72

ø

0.72

ğ

0.72

ă

0.72

č

0.72

Ė

0.72

ġ

0.71

Activations Density 1.105%

phrases or words related to numerical values and statistics

key phrases and indicators of trends or changes over time

No Comments

No Known Activations