INDEX

Explanations

words ending in 'ing' that are related to actions or behaviors

oai_token-act-pair · gpt-3.5-turbo

gerunds and present participles in verb forms

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

 blot

-0.60

 marking

-0.59

 breeze

-0.59

 grabs

-0.58

 consolidation

-0.58

 preview

-0.57

hurst

-0.57

GHC

-0.56

 precedence

-0.56

cut

-0.55

POSITIVE LOGITS

leep

0.81

ean

0.78

itely

0.78

yss

0.76

aciously

0.74

 peacefully

0.73

igmatic

0.72

iate

0.68

 respons

0.66

ibling

0.66

Activations Density 0.132%

words ending in 'ing' that are related to actions or behaviors

gerunds and present participles in verb forms

No Comments

No Known Activations