INDEX

Explanations

verbs related to negative actions and states

oai_token-act-pair · gpt-3.5-turbo

words that start with "dis" or relate to negative or detrimental concepts

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

 Suit

-0.74

 Juliet

-0.72

 Goth

-0.67

 Shogun

-0.66

OPLE

-0.66

 Pose

-0.65

 Scotia

-0.64

 Reach

-0.64

akeru

-0.63

 Mothers

-0.63

POSITIVE LOGITS

rup

1.26

burse

1.13

ordered

1.13

section

1.08

array

1.08

assemb

1.08

associated

1.08

organized

1.07

quiet

1.07

affected

1.06

Activations Density 0.012%

verbs related to negative actions and states

words that start with "dis" or relate to negative or detrimental concepts

No Comments

No Known Activations