INDEX

Explanations

phrases related to negative self-criticism or self-doubt

oai_token-act-pair · gpt-3.5-turbo

words involving self-destructive behaviors or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Assembly

-0.71

 guiIcon

-0.67

":["

-0.62

 Dhabi

-0.62

owan

-0.62

£ı

-0.60

iae

-0.60

MENTS

-0.59

Pod

-0.59

aviour

-0.58

POSITIVE LOGITS

hetically

0.80

 fuse

0.76

ocks

0.74

 crap

0.70

up

0.70

up

0.70

UP

0.68

ock

0.68

Up

0.67

 horn

0.67

Activations Density 0.479%

phrases related to negative self-criticism or self-doubt

words involving self-destructive behaviors or actions

No Comments

No Known Activations