INDEX

Explanations

the word "this" in contexts where it precedes a positive or noteworthy description or event

oai_token-act-pair · gpt-3.5-turbo

expressions of "this" referring to specific concepts or entities within a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ãĤ¹ãĥĪ

-0.80

aus

-0.79

ivas

-0.79

ARS

-0.77

Īè

-0.76

ãĥĨ

-0.74

ages

-0.73

acers

-0.72

¨

-0.71

masters

-0.70

POSITIVE LOGITS

 sucker

0.95

 trope

0.94

 week

0.91

guy

0.88

 particular

0.87

 nifty

0.83

 kind

0.80

 sort

0.79

 month

0.79

 type

0.79

Activations Density 0.198%

the word "this" in contexts where it precedes a positive or noteworthy description or event

expressions of "this" referring to specific concepts or entities within a narrative

No Comments

No Known Activations