INDEX

Explanations

phrases related to direct interpersonal interactions

oai_token-act-pair · gpt-3.5-turbo

phrases related to direct interactions or engagements

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

Ĥª

-0.78

 reluct

-0.71

aditional

-0.67

©¶æ¥µ

-0.67

oun

-0.63

 tremend

-0.63

senal

-0.59

 notor

-0.59

Þ

-0.57

tyr

-0.56

POSITIVE LOGITS

pped

1.11

ilet

1.08

pping

1.05

0.99

dden

0.93

oth

0.89

-$

0.86

plane

0.84

othy

0.84

KEN

0.80

Activations Density 0.020%

phrases related to direct interpersonal interactions

phrases related to direct interactions or engagements

No Comments

No Known Activations