INDEX

Explanations

words associated with negative connotations, including criticism and harm

oai_token-act-pair · gpt-3.5-turbo

words related to criticism and negative descriptors

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

minster

-0.75

 Gardens

-0.68

gow

-0.67

united

-0.64

bnb

-0.64

 laun

-0.63

aez

-0.61

etter

-0.60

soDeliveryDate

-0.60

urat

-0.59

POSITIVE LOGITS

ible

0.77

alion

0.76

chio

0.74

tein

0.74

dden

0.73

ction

0.68

fusc

0.68

ione

0.67

iblical

0.66

ected

0.65

Activations Density 0.054%

words associated with negative connotations, including criticism and harm

words related to criticism and negative descriptors

No Comments

No Known Activations