INDEX

Explanations

locations and numbers

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "the" indicating common nouns and locations within the text

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

ledge

-0.90

âĦ¢:

-0.80

BILITIES

-0.72

Figure

-0.72

ACA

-0.72

ucks

-0.72

BOOK

-0.71

ADA

-0.66

CLE

-0.66

Reviewer

-0.65

POSITIVE LOGITS

 wake

1.23

 meantime

1.22

 aftermath

1.21

 vicinity

1.19

 afternoon

1.07

 midst

1.07

wee

1.06

 past

1.04

 weeks

1.04

 evenings

1.03

Activations Density 0.154%

locations and numbers

the presence of the word "the" indicating common nouns and locations within the text

No Comments

No Known Activations