INDEX

Explanations

instances where something or someone is observed or seen

oai_token-act-pair · gpt-3.5-turbo

instances of the word "spotted."

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

vernment

-0.78

nesia

-0.77

industrial

-0.75

ayers

-0.70

bern

-0.68

idem

-0.68

mort

-0.67

pires

-0.66

 ancest

-0.66

aceutical

-0.66

POSITIVE LOGITS

owl

0.85

bugs

0.85

lights

0.82

 spotted

0.79

 sightings

0.78

Owl

0.74

 Viper

0.73

ãĤĬ

0.72

 glimps

0.71

flies

0.67

Activations Density 0.025%

instances where something or someone is observed or seen

instances of the word "spotted."

No Comments

No Known Activations