INDEX

Explanations

words related to extreme negativity or criticism

oai_token-act-pair · gpt-3.5-turbo

the repeated use of the word "utter" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

llah

-0.75

actionDate

-0.75

amsung

-0.75

xual

-0.74

OHN

-0.72

pei

-0.69

deen

-0.68

 Feder

-0.67

ysc

-0.66

dule

-0.66

POSITIVE LOGITS

ances

1.16

ance

0.99

most

0.89

aneous

0.83

TY

0.80

 utter

0.79

 amaz

0.75

ables

0.73

ing

0.73

ancing

0.73

Activations Density 0.016%

words related to extreme negativity or criticism

the repeated use of the word "utter" in various contexts

No Comments

No Known Activations