INDEX

Explanations

action verbs indicating movement

oai_token-act-pair · gpt-3.5-turbo

keywords related to positive experiences and enjoyment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

cknow

-0.76

inyl

-0.76

Failure

-0.73

iphate

-0.71

sequent

-0.71

failed

-0.70

cible

-0.69

threat

-0.68

vernment

-0.67

ourn

-0.67

POSITIVE LOGITS

 awesome

1.60

 amazing

1.60

 FANT

1.59

 fantastic

1.53

 wonderful

1.44

 HUGE

1.44

 GREAT

1.40

 lovely

1.36

 hilarious

1.34

 nice

1.33

Activations Density 0.475%

action verbs indicating movement

keywords related to positive experiences and enjoyment

No Comments

No Known Activations