INDEX

Explanations

adjectives or verbs indicating visibility or appearance

oai_token-act-pair · gpt-3.5-turbo

references to visibility or the concept of being observable

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

axter

-0.84

ework

-0.82

inion

-0.79

orously

-0.78

req

-0.77

annis

-0.77

rica

-0.76

isol

-0.75

ese

-0.74

upuncture

-0.73

POSITIVE LOGITS

 signs

0.85

 visible

0.83

 Signs

0.77

ibility

0.72

seeing

0.71

terday

0.68

lights

0.67

 silhouette

0.67

isable

0.66

 srfAttach

0.66

Activations Density 0.021%

adjectives or verbs indicating visibility or appearance

references to visibility or the concept of being observable

No Comments

No Known Activations