INDEX

Explanations

references to individuals or groups with specific attributes or roles

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ards."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONSORED

-0.70

ĸļ

-0.65

ĺħ

-0.65

gd

-0.65

emb

-0.62

 Podesta

-0.61

wei

-0.61

trak

-0.60

 amen

-0.57

mington

-0.57

POSITIVE LOGITS

ards

1.12

hips

0.94

ragon

0.89

ystem

0.86

hire

0.86

creen

0.86

inas

0.83

hip

0.82

ard

0.80

nesses

0.79

Activations Density 0.012%

references to individuals or groups with specific attributes or roles

occurrences of the word "ards."

No Comments

No Known Activations